机器学习工程师纳米学位¶
毕业项目：DeepTesla¶
2017年11月11日
说明：这个Jupter Notebook 文件包含了项目实现的代码以及相关注释。按照设计思路，将整个项目分为四个部分，分诉如下：
- Part 1: 数据分析与探索。这一步骤主要是读取数据（包括csv文件和视频文件）、分析数据（如转向角度的数据分布等）。
- Part 2: 数据预处理。在这部分中将前九段视频数据（转化为图片后）和对应转向角度及图片路径（转化为csv文件）作为训练数据（将划分为训练集和验证集）存储到硬盘中，将第十段视频（转化为图片后）和对应转向角度及图片路径（转化为csv文件）作为测试集存储到硬盘中。对训练数据和测试数据运用数据简化技术：剪切及重设尺寸；使用数据增加技术（之后在每个模型生成前，使用python data generator（生成器）来进行数据的分批次从硬盘导入，同时使用数据增加的方法。）：水平翻转和增加噪音等。尝试使用VAE+GAN增加转向角度较大的图像作为训练数据。
- Part 3: 利用Keras和Tensorflow训练模型并导出模型参数。使用基准模型对训练集进行训练，并通过验证集和测试集评测模型表现。期间涉及到技术有：卷积网络和全连接网络搭建与调参，迁移学习网络搭建与参数固定和调参，RELU和ELU激活函数运用，运用Droupout层防止过拟合等。最后将训练好模型和参数保存为model.json和model.h5两个文件。
- Part 4: 选择最佳模型并生成最终视频。使用时间序列图对比各模型表现，选择最终模型。并使用最终模型在测试视频epoch10_front.mkv上测试，评估人工转向与模型转向误差。最后对测试视频epoch10_front.mkv生成结果视频并保存在./output中。
Part 1: 数据分析与探索
导入相关库文件
In [1]:
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import matplotlib.image as mpimg
import os
import cv2
import glob
import random
%matplotlib inline
1.视频帧初探
In [4]:
### 载入前置摄像头图像
frame_capture = cv2.VideoCapture('./epochs/epoch01_front.mkv')
ret, img_test = frame_capture.read()
frame_capture.release()
plt.imshow(img_test)
Out[4]:
<matplotlib.image.AxesImage at 0x297d3cfa0f0>
In [5]:
### 查看图片尺寸
img_test.shape
Out[5]:
(720, 1280, 3)
In [11]:
### 视频不同场景截图
# 读取所有视频截图图片（不同场景）并显示出来
test_images = glob.glob('./images/Video_Snapshots/*.jpg')
fig, axs = plt.subplots(3, 3, figsize=(16,14))
fig.subplots_adjust(hspace = .004, wspace=.002)
axs = axs.ravel()
for i, im in enumerate(test_images):
axs[i].imshow(mpimg.imread(im))
axs[i].set_title('video snapshots'+str(i+1)+'.jpg', fontsize=20)
axs[i].axis('off')
# 显示所有图片
plt.tight_layout()
plt.show()
2.转向角度分析
In [13]:
### 加载所有转向角度csv文件
csv_files = glob.glob('./epochs/*steering.csv')
csv_steers = pd.concat((pd.read_csv(f) for f in csv_files))
csv_steers.head()
Out[13]:
| frame | frame_index | ts_micro | wheel | |
|---|---|---|---|---|
| 0 | 0.0 | NaN | 1464650070285914 | -1.0 |
| 1 | 1.0 | NaN | 1464650070319247 | -1.0 |
| 2 | 2.0 | NaN | 1464650070352581 | -1.0 |
| 3 | 3.0 | NaN | 1464650070385914 | -1.0 |
| 4 | 4.0 | NaN | 1464650070419247 | -1.0 |
In [14]:
### 查看转向角度统计信息
csv_steers.describe()
Out[14]:
| frame | frame_index | ts_micro | wheel | |
|---|---|---|---|---|
| count | 5400.000000 | 21600.000000 | 2.700000e+04 | 27000.000000 |
| mean | 1616.166667 | 1349.500000 | 1.464374e+15 | -0.338667 |
| std | 1120.991308 | 779.440853 | 1.382617e+11 | 4.438301 |
| min | 0.000000 | 0.000000 | 1.464304e+15 | -18.000000 |
| 25% | 674.750000 | 674.750000 | 1.464304e+15 | -2.500000 |
| 50% | 1349.500000 | 1349.500000 | 1.464305e+15 | 0.000000 |
| 75% | 2549.250000 | 2024.250000 | 1.464306e+15 | 1.500000 |
| max | 3899.000000 | 2699.000000 | 1.464650e+15 | 15.000000 |
In [15]:
### 直方图查看'wheel'方向角度数据分布
wheel = csv_steers['wheel']
plt.figure
plt.grid(True)
plt.hist(wheel,100);
plt.xlabel('wheel')
plt.ylabel('counts')
plt.savefig('./images/img/Hist.png')
In [25]:
### 查看转向角度的时间序列图
#生成时间序列图(2700帧，90秒)
plt.figure;
plt.plot(np.arange( 0, 2700),wheel[:2700]);
plt.xlabel('Frame Counts')
plt.ylabel('Steering Angles')
plt.grid(True)
plt.savefig('./images/img/Frame_angles.png')
plt.show()
3.视频帧和转向角度结合
In [27]:
### 读取'epoch03_front.mkv'和对应'epoch03_steering.csv'文件，
frame_capture3 = cv2.VideoCapture('./epochs/epoch03_front.mkv')
# 抓取到内存中
epoch3_imgs = []
while True:
ret, img = frame_capture3.read()
if not ret:
break
epoch3_imgs.append(img)
print ("The epoch03 images' length is:",len(epoch3_imgs))
# 读入对应csv文件
csv_wheel3 = pd.read_csv('./epochs/epoch03_steering.csv')
steering_wheel3 = csv_wheel3['wheel'].values
frame_wheel3 = csv_wheel3['frame_index'].values
print("The epoch03 wheel length is", len(steering_wheel3))
The epoch03 images' length is: 2700 The epoch03 wheel length is 2700
In [28]:
### 随机查看帧图片和转向策略
# 随机查看9张图片和转向角度
fig, axs = plt.subplots(3, 3, figsize=(16,14))
fig.subplots_adjust(hspace = 0, wspace=0)
axs = axs.ravel()
for i in range(9):
index = random.choice(range(len(epoch3_imgs)))
img = epoch3_imgs[index]
#ax = fig.add_subplot(3,3,i+1)
axs[i].imshow(img)
axs[i].set_title('frame:'+ str(frame_wheel3[index]) + ' wheel:' + str(steering_wheel3[index]), fontsize=20)
# 显示所有图片
plt.tight_layout()
plt.savefig('./images/img/frame_wheel.png')
plt.show()
Part 2: 数据预处理
导入相关库文件
说明：每个part开始时，重启Kernel
In [3]:
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import cv2
import random
import os
import csv
import tensorflow as tf
import pickle
# 导入自定义数据处理文件'preprocess_data'
import preprocess_data
# 导入自定义'vae_gan'
import vae_gan
z_dim = 512
VG_PATH = './vg_images/'
%matplotlib inline
Using TensorFlow backend.
1.单张图片处理测试
In [22]:
### <1>读取测试图片
image_test = cv2.imread('./images/img/frame_1173.jpg')
image_test = cv2.cvtColor(image_test,cv2.COLOR_BGR2RGB)
print(image_test.shape)
plt.imshow(image_test)
(720, 1280, 3)
Out[22]:
<matplotlib.image.AxesImage at 0x206bc3415f8>
In [23]:
### <2>裁剪掉天空部分，并重设图片尺寸大小为80*80，后续用于数据选取简化（可加快模型收敛速度）
image_test_crop = image_test[300:720, 0:1280]
image_test_resize = cv2.resize(image_test_crop, (80, 80))
print(image_test_resize.shape)
plt.imshow(image_test_resize)
(80, 80, 3)
Out[23]:
<matplotlib.image.AxesImage at 0x206bb86d4a8>
In [24]:
### <3>水平翻转图片，图片数据归一化，并给图片增加噪声，后续用于数据增加
#翻转图片
image_test_flip = cv2.flip(image_test_resize, 1)
#图片数据归一化
image_test_flip = 0.1 + (image_test_flip - 0)*(0.9 - 0.1) / 255.0
#增加噪声
noisy_img = image_test_flip + 0.05 * np.random.randn(*image_test_flip.shape)
noisy_img = np.clip(noisy_img, 0.1, 0.9)
plt.imshow(noisy_img)
Out[24]:
<matplotlib.image.AxesImage at 0x206bb8d40f0>
2.原始训练数据与测试数据生成
In [2]:
### <1>使用 'preprocess_data' 中’load_data'将所有mkv文件转化为图像(rgb jpg格式)装入内存返回和在硬盘持久化
###(分别保存在./train_images和./test_images文件夹中)
### 并将所有对应转向角度和图片路径装入内存返回和在硬盘持久化(分别保存在./train_images和./test_images中csv文件)
#使用"1.单张图片处理测试"中的数据简化（裁剪，重设尺寸）生成基本数据
train_imgs, train_wheels = preprocess_data.load_data('train',write_to_disk=True)
test_imgs, test_wheels = preprocess_data.load_data('test',write_to_disk=True)
load data start! The epoch1 mkv is processing The epoch2 mkv is processing The epoch3 mkv is processing The epoch4 mkv is processing The epoch5 mkv is processing The epoch6 mkv is processing The epoch7 mkv is processing The epoch8 mkv is processing The epoch9 mkv is processing loading data filished! load data start! The epoch10 mkv is processing loading data filished!
In [3]:
### <2>测试生成的训练和测试图片的shape
print("The shape of train images is", train_imgs.shape)
print("The shape of train wheels", train_wheels.shape)
print("The shape of test images is", test_imgs.shape)
print("The shape of test wheels", test_wheels.shape)
The shape of train images is (24300, 80, 80, 3) The shape of train wheels (24300, 1) The shape of test images is (2700, 80, 80, 3) The shape of test wheels (2700, 1)
In [4]:
### <3>测试保存在硬盘上的图片和csv文件
f, (ax1,ax2) = plt.subplots(1, 2, figsize=(20, 10))
f.tight_layout()
img_train = plt.imread('./train_images/epoch01_steering_0.jpg')
csv_train = pd.read_csv('./train_images/train_images.csv')
ax1.imshow(img_train)
ax1.set_title(csv_train['img_path'][0]+' wheel:'+str(csv_train['wheel'][0]), fontsize=15)
img_test = plt.imread('./test_images/epoch10_steering_0.jpg')
csv_test = pd.read_csv('./test_images/test_images.csv')
ax2.imshow(img_test)
ax2.set_title(csv_test['img_path'][0]+' wheel:'+str(csv_test['wheel'][0]), fontsize=15)
Out[4]:
<matplotlib.text.Text at 0x279c17ad898>
In [1]:
### <4>对整体的数据处理策略为：
### 使用 python data generator（生成器）来进行数据的分批次从硬盘导入，同时使用数据增加的方法，对每批次进行数据增加处理（翻转和增加噪音）
### 在使用完数据增加后，数据总体将是原来的3倍
### 具体代码见文件'./preprocess_data.py中的'generator'函数
import preprocess_data
from sklearn.model_selection import train_test_split
#载入训练数据
#之后每个模型单独进行 训练数据和验证数据生成
samples = preprocess_data.read_csv('./train_images/train_images.csv')
train_samples, validation_samples = train_test_split(samples, test_size=0.2)
train_generator = preprocess_data.generator(train_samples, batch_size=128)
validation_generator = preprocess_data.generator(validation_samples, batch_size=128)
In [2]:
# 测试'generator'
my_output = []
for i in range(10):
my_output = (next(train_generator))
print(my_output[0].shape)
print(my_output[1].shape)
(384, 80, 80, 3) (384,)
3.使用 GAN+VAE 生成训练用数据(选做)
说明：
模型建立文章来源于：Learning a Driving Simulator中的Autoencoder部分
参考实现代码：https://github.com/commaai/research 其中的train_generative_model.py及./models/autoencoder.py, layer.py
具体实现见本项目文件中./vae_gan.py文件中
In [4]:
### <1>建立存储中间过程文件夹
if not os.path.exists("./outputs/results_"+"autoencoder"):
os.makedirs("./outputs/results_"+"autoencoder")
if not os.path.exists("./outputs/samples_"+"autoencoder"):
os.makedirs("./outputs/samples_"+"autoencoder")
In [5]:
### <2>定义模型的训练生成器
### 提取转向角度小于-5或大于5的图片为训练数据（从第一部分直方图查看'wheel'方向角度数据分布中可以看出，这些数据较少）
def read_csv():
"""
# Reading the csv file and store in the list
"""
samples_original = []
samples_steering = []
with open('./train_images/train_images.csv') as csvfile:
reader = csv.reader(csvfile)
for line in reader:
samples_original.append(line)
samples_original = samples_original[1:]
# 提取转向角度小于-5或大于5的sample
for sample in samples_original:
if (float(sample[0]) < -5.0) or (float(sample[0]) > 5.0):
samples_steering.append(sample)
#增加转向幅度大的sample
while True:
if len(samples_steering) < 8000:
#print(len(samples_steering))
index = random.choice(range(len(samples_steering)))
samples_steering.append(samples_steering[index])
else:
break
return samples_steering
def generator_train(samples, batch_size=64):
"""
# The generator of the samples
"""
num_samples = len(samples)
while True: #循环不停生成
for offset in range(0, num_samples, batch_size):
batch_samples = samples[offset:offset+batch_size]
X_train = []
for sample in batch_samples:
X = cv2.imread(sample[1])
X = cv2.resize(X, (160, 80))
X = cv2.cvtColor(X,cv2.COLOR_BGR2RGB)
X = X/127.5 - 1.
X_train.append(X)
#这里是无监督学习，不使用wheels, 使用Z作为随机输入
X_train = np.array(X_train)
Z = np.random.normal(0, 1, (X_train.shape[0], z_dim))
yield Z, X_train
# 读取csv文件每一行作为一个sample
train_samles = read_csv()
#train_samles = train_samles[0:8000]
print(np.array(train_samles).shape)
# 测试'generator_train'
my_output = []
for i in range(10):
my_output = (next(generator_train(train_samles)))
print(my_output[0].shape)
print(my_output[1].shape)
(8000, 2) (64, 512) (64, 80, 160, 3)
In [6]:
### <3>模型训练生成图片
with tf.Session() as sess:
g_train, d_train, sampler, saver, loader, extras = vae_gan.get_model(sess=sess, name="autoencoder", batch_size=64, gpu=0)
sampler_train = vae_gan.train_model("autoencoder", g_train, d_train, sampler,
generator_train(train_samles),
samples_per_epoch=8000,
nb_epoch=20, verbose=1, saver=saver
)
# 训练完后使用生成图片
new_samples = []
for i in range(8000//64):
output = (next(generator_train(train_samles)))
new_sample, _ = sampler(output[0], output[1])
new_samples.append(new_sample)
D:\Anaconda3\envs\carnd-term1\lib\site-packages\keras\engine\topology.py:379: UserWarning: The `regularizers` property of layers/models is deprecated. Regularization losses are now managed via the `losses` layer/model property.
warnings.warn('The `regularizers` property of layers/models '
G.shape: (64, 80, 160, 3) E.shape: [(64, 512), (64, 512)] D.shape: [(64, 1), (64, 5, 10, 512)] Generator variables: autoencoder/g_h0_lin_W:0 autoencoder/g_h0_lin_b:0 autoencoder/g_bn0_gamma:0 autoencoder/g_bn0_beta:0 autoencoder/g_h1/w:0 autoencoder/g_h1/biases:0 autoencoder/g_bn1_gamma:0 autoencoder/g_bn1_beta:0 autoencoder/g_h2/w:0 autoencoder/g_h2/biases:0 autoencoder/g_bn2_gamma:0 autoencoder/g_bn2_beta:0 autoencoder/g_h3/w:0 autoencoder/g_h3/biases:0 autoencoder/g_bn3_gamma:0 autoencoder/g_bn3_beta:0 autoencoder/g_h4/w:0 autoencoder/g_h4/biases:0 Discriminator variables: autoencoder/d_h0_conv_W:0 autoencoder/d_h0_conv_b:0 autoencoder/d_h1_conv_W:0 autoencoder/d_h1_conv_b:0 autoencoder/d_bn1_gamma:0 autoencoder/d_bn1_beta:0 autoencoder/d_h2_conv_W:0 autoencoder/d_h2_conv_b:0 autoencoder/d_bn2_gamma:0 autoencoder/d_bn2_beta:0 autoencoder/d_h3_conv_W:0 autoencoder/d_h3_conv_b:0 autoencoder/d_bn3_gamma:0 autoencoder/d_bn3_beta:0 autoencoder/d_h3_lin_W:0 autoencoder/d_h3_lin_b:0 Encoder variables: autoencoder/e_h0_conv_W:0 autoencoder/e_h0_conv_b:0 autoencoder/e_h1_conv_W:0 autoencoder/e_h1_conv_b:0 autoencoder/e_bn1_gamma:0 autoencoder/e_bn1_beta:0 autoencoder/e_h2_conv_W:0 autoencoder/e_h2_conv_b:0 autoencoder/e_bn2_gamma:0 autoencoder/e_bn2_beta:0 autoencoder/e_h3_conv_W:0 autoencoder/e_h3_conv_b:0 autoencoder/e_bn3_gamma:0 autoencoder/e_bn3_beta:0 autoencoder/e_h3_lin_W:0 autoencoder/e_h3_lin_b:0 autoencoder/e_h4_lin_W:0 autoencoder/e_h4_lin_b:0 Epoch 1/20 8000/8000 [==============================] - 252s - g_loss: 1.8224 - d_loss: 1.8719 - d_loss_fake: 1.1442 - d_loss_legit: 0.7277 - time: 1.8067 Epoch 2/20 8000/8000 [==============================] - 249s - g_loss: 1.9424 - d_loss: 2.7553 - d_loss_fake: 1.4145 - d_loss_legit: 1.3408 - time: 1.7958 Epoch 3/20 8000/8000 [==============================] - 249s - g_loss: 2.4958 - d_loss: 2.5902 - d_loss_fake: 1.2287 - d_loss_legit: 1.3615 - time: 1.7970 Epoch 4/20 8000/8000 [==============================] - 249s - g_loss: 2.7573 - d_loss: 2.1233 - d_loss_fake: 0.8719 - d_loss_legit: 1.2514 - time: 1.7951 Epoch 5/20 8000/8000 [==============================] - 253s - g_loss: 2.9694 - d_loss: 1.7823 - d_loss_fake: 0.7848 - d_loss_legit: 0.9975 - time: 1.8014 Epoch 6/20 8000/8000 [==============================] - 252s - g_loss: 3.1063 - d_loss: 1.7980 - d_loss_fake: 0.7422 - d_loss_legit: 1.0557 - time: 1.8025 Epoch 7/20 8000/8000 [==============================] - 252s - g_loss: 2.9509 - d_loss: 1.7951 - d_loss_fake: 0.8246 - d_loss_legit: 0.9705 - time: 1.8015 Epoch 8/20 8000/8000 [==============================] - 253s - g_loss: 3.1742 - d_loss: 1.7414 - d_loss_fake: 0.7171 - d_loss_legit: 1.0244 - time: 1.8048 Epoch 9/20 8000/8000 [==============================] - 251s - g_loss: 2.8747 - d_loss: 1.5559 - d_loss_fake: 0.6708 - d_loss_legit: 0.8850 - time: 1.8055 Epoch 10/20 8000/8000 [==============================] - 252s - g_loss: 2.9815 - d_loss: 1.5117 - d_loss_fake: 0.6822 - d_loss_legit: 0.8295 - time: 1.8043 Epoch 11/20 8000/8000 [==============================] - 250s - g_loss: 2.9477 - d_loss: 1.3911 - d_loss_fake: 0.5788 - d_loss_legit: 0.8123 - time: 1.8034 Epoch 12/20 8000/8000 [==============================] - 250s - g_loss: 2.7813 - d_loss: 1.3522 - d_loss_fake: 0.5656 - d_loss_legit: 0.7866 - time: 1.8010 Epoch 13/20 8000/8000 [==============================] - 250s - g_loss: 2.9709 - d_loss: 1.3922 - d_loss_fake: 0.6117 - d_loss_legit: 0.7805 - time: 1.8029 Epoch 14/20 8000/8000 [==============================] - 251s - g_loss: 2.9065 - d_loss: 1.2279 - d_loss_fake: 0.5358 - d_loss_legit: 0.6920 - time: 1.7979 Epoch 15/20 8000/8000 [==============================] - 252s - g_loss: 3.2203 - d_loss: 1.2404 - d_loss_fake: 0.5362 - d_loss_legit: 0.7042 - time: 1.8021 Epoch 16/20 8000/8000 [==============================] - 252s - g_loss: 3.0949 - d_loss: 1.2104 - d_loss_fake: 0.5228 - d_loss_legit: 0.6876 - time: 1.8023 Epoch 17/20 8000/8000 [==============================] - 251s - g_loss: 3.0838 - d_loss: 1.1478 - d_loss_fake: 0.4915 - d_loss_legit: 0.6563 - time: 1.8037 Epoch 18/20 8000/8000 [==============================] - 252s - g_loss: 3.2523 - d_loss: 1.0994 - d_loss_fake: 0.4588 - d_loss_legit: 0.6405 - time: 1.8057 Epoch 19/20 8000/8000 [==============================] - 251s - g_loss: 3.0428 - d_loss: 1.1564 - d_loss_fake: 0.5095 - d_loss_legit: 0.6469 - time: 1.8046 Epoch 20/20 8000/8000 [==============================] - 251s - g_loss: 3.1535 - d_loss: 0.9698 - d_loss_fake: 0.4197 - d_loss_legit: 0.5501 - time: 1.8046
In [7]:
### <4>保存和查看生成图片
### 图中奇数列（从1开始计数）为生成图片，偶数列（从2开始）为原始图片。生成图片基本呈现了原始图片特征并有细微差别（转向角度大体没变），
### 后面尝试使用训练
### 原始模型进行了200次训练，这里因时间原因和练习目的仅训练20次。后期可增加训练次数进行优化
# 保存新生成图像和相关信息
pickle.dump(new_samples, open('./images/vg_images.p', 'wb'))
pickle.dump(train_samles, open('./images/train_samles.p', 'wb'))
#查看新生成图片
image_test = cv2.imread('./outputs/samples_autoencoder/train_19_0.png')
image_test = cv2.cvtColor(image_test,cv2.COLOR_BGR2RGB)
plt.figure(figsize=(15,15))
plt.imshow(image_test)
Out[7]:
<matplotlib.image.AxesImage at 0x25e8b8667f0>
In [8]:
### <5>载入新生成图片和相关信息并测试
# 载入数据
imgs_data = pickle.load(open('./images/vg_images.p', 'rb'))
wheels_data = pickle.load(open('./images/train_samles.p', 'rb'))
# 查看图片
new_images = np.array(imgs_data).reshape((-1,80,160,3))
print(new_images.shape)
print(np.array(wheels_data).shape)
new_images = vae_gan.inverse_transform(new_images)
plt.imshow(new_images[799])
(8000, 80, 160, 3) (8000, 2)
Out[8]:
<matplotlib.image.AxesImage at 0x25e6da7b940>
In [9]:
### <6>将图片(以rgb jpg格式)和转向角度及图片路径(以csv文件)保存到硬盘'./vg_images'文件夹中
preprocess_data.write_vg_images(new_images, wheels_data)
Write vg images success!
In [10]:
### <7>测试保存在硬盘上新生成csv文件和原文件对比
f, (ax1,ax2) = plt.subplots(1, 2, figsize=(15, 10))
f.tight_layout()
img_vg = plt.imread('./vg_images/vg_images0.jpg')
csv_vg = pd.read_csv('./vg_images/vg_images.csv')
ax1.imshow(img_vg)
ax1.set_title('VG IMAGE: '+csv_vg['img_path'][0]+' wheel:'+str(csv_vg['wheel'][0]), fontsize=15)
img_orginal = plt.imread(wheels_data[0][1])
ax2.imshow(img_orginal)
ax2.set_title('ORIGINAL: '+wheels_data[0][1]+' wheel:'+wheels_data[0][0], fontsize=15)
Out[10]:
<matplotlib.text.Text at 0x25e6661fdd8>
数据处理方式说明：之后在每个模型生成前，使用python data generator（生成器）来进行数据的分批次从硬盘导入（先读取csv文件，再根据文件中的路径读取图片），同时使用数据增加的方法，对每批次进行数据增加处理。
Part 3: 训练模型
说明：每个model开始时，重启Kernel
1.NVIDIA end-to-end model
(1) Benchmark (基准模型)¶
In [1]:
### <1>用generator 生成NVIDIA model训练数据
import preprocess_data
from sklearn.model_selection import train_test_split
#载入训练数据
samples = preprocess_data.read_csv('./train_images/train_images.csv')
train_samples, validation_samples = train_test_split(samples, test_size=0.2)
train_generator = preprocess_data.generator(train_samples, batch_size=128)
validation_generator = preprocess_data.generator(validation_samples, batch_size=128)
In [2]:
### <2>建立NVIDIA model
from keras.models import Sequential
from keras.layers import Convolution2D, Dropout, Flatten, Dense, Lambda
def Nvidia_model():
"""
Nvidia model:referenced https://arxiv.org/pdf/1604.07316v1.pdf
"""
print("Nvidia-------------")
model = Sequential()
model.add(Convolution2D(24,5,5,subsample=(2,2),activation='relu', input_shape=(80,80,3)))
model.add(Convolution2D(36,5,5,subsample=(2,2),activation='relu'))
model.add(Convolution2D(48,5,5,subsample=(2,2),activation='relu'))
model.add(Convolution2D(64,3,3,activation='relu'))
model.add(Convolution2D(64,3,3,activation='relu'))
model.add(Flatten())
model.add(Dense(1164, activation='relu'))
model.add(Dense(100, activation='relu'))
model.add(Dense(50, activation='relu'))
model.add(Dense(10, activation='relu'))
model.add(Dense(1))
return model
Using TensorFlow backend.
In [3]:
Nvidia_model().summary()
Nvidia------------- ____________________________________________________________________________________________________ Layer (type) Output Shape Param # Connected to ==================================================================================================== convolution2d_1 (Convolution2D) (None, 38, 38, 24) 1824 convolution2d_input_1[0][0] ____________________________________________________________________________________________________ convolution2d_2 (Convolution2D) (None, 17, 17, 36) 21636 convolution2d_1[0][0] ____________________________________________________________________________________________________ convolution2d_3 (Convolution2D) (None, 7, 7, 48) 43248 convolution2d_2[0][0] ____________________________________________________________________________________________________ convolution2d_4 (Convolution2D) (None, 5, 5, 64) 27712 convolution2d_3[0][0] ____________________________________________________________________________________________________ convolution2d_5 (Convolution2D) (None, 3, 3, 64) 36928 convolution2d_4[0][0] ____________________________________________________________________________________________________ flatten_1 (Flatten) (None, 576) 0 convolution2d_5[0][0] ____________________________________________________________________________________________________ dense_1 (Dense) (None, 1164) 671628 flatten_1[0][0] ____________________________________________________________________________________________________ dense_2 (Dense) (None, 100) 116500 dense_1[0][0] ____________________________________________________________________________________________________ dense_3 (Dense) (None, 50) 5050 dense_2[0][0] ____________________________________________________________________________________________________ dense_4 (Dense) (None, 10) 510 dense_3[0][0] ____________________________________________________________________________________________________ dense_5 (Dense) (None, 1) 11 dense_4[0][0] ==================================================================================================== Total params: 925,047 Trainable params: 925,047 Non-trainable params: 0 ____________________________________________________________________________________________________
In [4]:
### <3>NVIDIA model模型训练
model = Nvidia_model()
model.compile(loss='mse', optimizer='adam')
history_object = model.fit_generator(train_generator, samples_per_epoch=
len(train_samples)*3, validation_data=validation_generator,
nb_val_samples=len(validation_samples)*3,nb_epoch=10)
Nvidia------------- Epoch 1/10 58320/58320 [==============================] - 41s - loss: 17.8832 - val_loss: 10.7253 Epoch 2/10 58320/58320 [==============================] - 38s - loss: 7.3714 - val_loss: 4.8491 Epoch 3/10 58320/58320 [==============================] - 36s - loss: 3.8005 - val_loss: 2.4994 Epoch 4/10 58320/58320 [==============================] - 37s - loss: 2.1176 - val_loss: 1.5325 Epoch 5/10 58320/58320 [==============================] - 37s - loss: 1.3325 - val_loss: 1.1606 Epoch 6/10 58320/58320 [==============================] - 37s - loss: 1.0151 - val_loss: 0.8715 Epoch 7/10 58320/58320 [==============================] - 38s - loss: 0.8035 - val_loss: 0.8606 Epoch 8/10 58320/58320 [==============================] - 38s - loss: 0.6615 - val_loss: 0.7174 Epoch 9/10 58320/58320 [==============================] - 37s - loss: 0.5866 - val_loss: 0.5787 Epoch 10/10 58320/58320 [==============================] - 36s - loss: 0.5361 - val_loss: 0.5013
In [5]:
### <4>分析结果
import preprocess_data
import matplotlib.pyplot as plt
#读取测试测试文件
test_imgs, test_wheels = preprocess_data.load_data('test')
test_imgs = preprocess_data.nomorlize_image(test_imgs)
# 测试模型在测试文件上表现
test_loss= model.evaluate(test_imgs, test_wheels, batch_size=128)
print('\n Test loss is:{}'.format(test_loss))
# 查看训练集和验证集loss
plt.plot(history_object.history['loss'])
plt.plot(history_object.history['val_loss'])
plt.title('Nvidia model mean squared error loss')
plt.ylabel('mean squared error loss')
plt.xlabel('epoch')
plt.legend(['training set', 'validation set'], loc='upper right')
plt.show()
load data start! The epoch10 mkv is processing loading data filished! 2700/2700 [==============================] - 1s Test loss is:8.861750322977702
In [6]:
### <5>保存model
import os
# model文件和json文件存储路径
model_saved_path = os.path.join('./models/', "nvidia_model.h5")
json_saved_path = os.path.join('./models/', "nvidia_model.json")
# 存储json
json_model = model.to_json()
with open(json_saved_path, "w") as json_file:
json_file.write(json_model)
# 存储model
model.save(model_saved_path)
In [14]:
### 选做：<6>增加vg+gan生成图片到训练数据
#载入训练数据
samples = preprocess_data.read_csv('./train_images/train_images.csv')
vg_samples = preprocess_data.read_csv('./vg_images/vg_images.csv')
train_samples, validation_samples = train_test_split(samples, test_size=0.2)
train_samples = train_samples + vg_samples
train_generator = preprocess_data.generator(train_samples, batch_size=128)
validation_generator = preprocess_data.generator(validation_samples, batch_size=128)
model_plus = Nvidia_model()
model_plus.compile(loss='mse', optimizer='adam')
history_object = model_plus.fit_generator(train_generator, samples_per_epoch=
len(train_samples)*3, validation_data=validation_generator,
nb_val_samples=len(validation_samples)*3,nb_epoch=5)
Nvidia------------- Epoch 1/5 82320/82320 [==============================] - 56s - loss: 38.1551 - val_loss: 20.3620 Epoch 2/5 82320/82320 [==============================] - 53s - loss: 40.7791 - val_loss: 20.3723 Epoch 3/5 82320/82320 [==============================] - 53s - loss: 40.5858 - val_loss: 20.3775 Epoch 4/5 82320/82320 [==============================] - 53s - loss: 41.5649 - val_loss: 20.3687 Epoch 5/5 82320/82320 [==============================] - 53s - loss: 41.4422 - val_loss: 20.3613
In [15]:
# 测试模型在测试文件上表现
test_loss= model_plus.evaluate(test_imgs, test_wheels, batch_size=128)
test_imgs = preprocess_data.nomorlize_image(test_imgs)
print('Test loss is:{}'.format(test_loss))
2688/2700 [============================>.] - ETA: 0sTest loss is:7.497045198016696
分析：在加入vae+gan模型生成数据后，原始模型表现在训练集和验证集不能很好收敛，但在测试集上表现变好（说明模型识能力有所增强）。但总体来说还是生成图片模型vae+gan训练次数太少，所以后期不加入vae+gan模型生成数据，留待后续优化。
(2) Refined Model (改进模型:ELU, Batch normalization, Weight initial, MaxPooling, BatchNormalization, Dropout)¶
In [1]:
### <1>用generator 生成NVIDIA refined model训练数据
import preprocess_data
from sklearn.model_selection import train_test_split
#载入训练数据
samples = preprocess_data.read_csv('./train_images/train_images.csv')
train_samples, validation_samples = train_test_split(samples, test_size=0.2)
train_generator = preprocess_data.generator(train_samples, batch_size=128)
validation_generator = preprocess_data.generator(validation_samples, batch_size=128)
In [2]:
### <2>建立NVIDIA refined model
from keras.models import Sequential
from keras.layers import Convolution2D, Dropout, Flatten, Dense, Lambda, BatchNormalization, MaxPooling2D
def Nvidia_refined_model():
"""
Nvidia refined model
"""
print("Nvidia-------------")
model = Sequential()
model.add(Convolution2D(24,5,5,subsample=(2,2),activation='elu',border_mode='same',init='he_normal',input_shape=(80,80,3)))
model.add(BatchNormalization())
model.add(MaxPooling2D())
model.add(Convolution2D(36,5,5,subsample=(2,2),activation='elu',border_mode='same',init='he_normal'))
model.add(BatchNormalization())
model.add(MaxPooling2D())
model.add(Convolution2D(48,5,5,subsample=(2,2),activation='elu',border_mode='same',init='he_normal'))
model.add(BatchNormalization())
model.add(MaxPooling2D())
model.add(Convolution2D(64,3,3,activation='elu',border_mode='same',init='he_normal'))
model.add(BatchNormalization())
model.add(Convolution2D(64,3,3,activation='elu',border_mode='same',init='he_normal'))
model.add(BatchNormalization())
model.add(Flatten())
model.add(Dense(1164, activation='elu',init='he_normal'))
model.add(Dropout(0.5))
model.add(Dense(100, activation='elu',init='he_normal'))
model.add(Dropout(0.3))
model.add(Dense(50, activation='elu',init='he_normal'))
model.add(Dropout(0.2))
model.add(Dense(10, activation='elu',init='he_normal'))
model.add(Dropout(0.1))
model.add(Dense(1))
return model
Using TensorFlow backend.
In [3]:
Nvidia_refined_model().summary()
Nvidia------------- ____________________________________________________________________________________________________ Layer (type) Output Shape Param # Connected to ==================================================================================================== convolution2d_1 (Convolution2D) (None, 40, 40, 24) 1824 convolution2d_input_1[0][0] ____________________________________________________________________________________________________ batchnormalization_1 (BatchNorma (None, 40, 40, 24) 96 convolution2d_1[0][0] ____________________________________________________________________________________________________ maxpooling2d_1 (MaxPooling2D) (None, 20, 20, 24) 0 batchnormalization_1[0][0] ____________________________________________________________________________________________________ convolution2d_2 (Convolution2D) (None, 10, 10, 36) 21636 maxpooling2d_1[0][0] ____________________________________________________________________________________________________ batchnormalization_2 (BatchNorma (None, 10, 10, 36) 144 convolution2d_2[0][0] ____________________________________________________________________________________________________ maxpooling2d_2 (MaxPooling2D) (None, 5, 5, 36) 0 batchnormalization_2[0][0] ____________________________________________________________________________________________________ convolution2d_3 (Convolution2D) (None, 3, 3, 48) 43248 maxpooling2d_2[0][0] ____________________________________________________________________________________________________ batchnormalization_3 (BatchNorma (None, 3, 3, 48) 192 convolution2d_3[0][0] ____________________________________________________________________________________________________ maxpooling2d_3 (MaxPooling2D) (None, 1, 1, 48) 0 batchnormalization_3[0][0] ____________________________________________________________________________________________________ convolution2d_4 (Convolution2D) (None, 1, 1, 64) 27712 maxpooling2d_3[0][0] ____________________________________________________________________________________________________ batchnormalization_4 (BatchNorma (None, 1, 1, 64) 256 convolution2d_4[0][0] ____________________________________________________________________________________________________ convolution2d_5 (Convolution2D) (None, 1, 1, 64) 36928 batchnormalization_4[0][0] ____________________________________________________________________________________________________ batchnormalization_5 (BatchNorma (None, 1, 1, 64) 256 convolution2d_5[0][0] ____________________________________________________________________________________________________ flatten_1 (Flatten) (None, 64) 0 batchnormalization_5[0][0] ____________________________________________________________________________________________________ dense_1 (Dense) (None, 1164) 75660 flatten_1[0][0] ____________________________________________________________________________________________________ dropout_1 (Dropout) (None, 1164) 0 dense_1[0][0] ____________________________________________________________________________________________________ dense_2 (Dense) (None, 100) 116500 dropout_1[0][0] ____________________________________________________________________________________________________ dropout_2 (Dropout) (None, 100) 0 dense_2[0][0] ____________________________________________________________________________________________________ dense_3 (Dense) (None, 50) 5050 dropout_2[0][0] ____________________________________________________________________________________________________ dropout_3 (Dropout) (None, 50) 0 dense_3[0][0] ____________________________________________________________________________________________________ dense_4 (Dense) (None, 10) 510 dropout_3[0][0] ____________________________________________________________________________________________________ dropout_4 (Dropout) (None, 10) 0 dense_4[0][0] ____________________________________________________________________________________________________ dense_5 (Dense) (None, 1) 11 dropout_4[0][0] ==================================================================================================== Total params: 330,023 Trainable params: 329,551 Non-trainable params: 472 ____________________________________________________________________________________________________
In [4]:
### <3>NVIDIA refined model模型训练
model = Nvidia_refined_model()
model.compile(loss='mse', optimizer='adam')
history_object = model.fit_generator(train_generator, samples_per_epoch=
len(train_samples)*3, validation_data=validation_generator,
nb_val_samples=len(validation_samples)*3,nb_epoch=10)
Nvidia------------- Epoch 1/10 58320/58320 [==============================] - 41s - loss: 12.3596 - val_loss: 16.3098 Epoch 2/10 58320/58320 [==============================] - 35s - loss: 4.8543 - val_loss: 7.3601 Epoch 3/10 58320/58320 [==============================] - 35s - loss: 3.2042 - val_loss: 2.5542 Epoch 4/10 58320/58320 [==============================] - 36s - loss: 2.5189 - val_loss: 1.3969 Epoch 5/10 58320/58320 [==============================] - 38s - loss: 2.1358 - val_loss: 1.1753 Epoch 6/10 58320/58320 [==============================] - 37s - loss: 1.8533 - val_loss: 1.0263 Epoch 7/10 58320/58320 [==============================] - 36s - loss: 1.7034 - val_loss: 1.3407 Epoch 8/10 58320/58320 [==============================] - 35s - loss: 1.5898 - val_loss: 0.8573 Epoch 9/10 58320/58320 [==============================] - 35s - loss: 1.4742 - val_loss: 0.7142 Epoch 10/10 58320/58320 [==============================] - 37s - loss: 1.4000 - val_loss: 0.8484
In [5]:
### <4>分析结果
import preprocess_data
import matplotlib.pyplot as plt
#读取测试测试文件
test_imgs, test_wheels = preprocess_data.load_data('test')
test_imgs = preprocess_data.nomorlize_image(test_imgs)
# 测试模型在测试文件上表现
test_loss= model.evaluate(test_imgs, test_wheels, batch_size=128)
print('\n Test loss is:{}'.format(test_loss))
# 查看训练集和验证集loss
plt.plot(history_object.history['loss'])
plt.plot(history_object.history['val_loss'])
plt.title('Nvidia model mean squared error loss')
plt.ylabel('mean squared error loss')
plt.xlabel('epoch')
plt.legend(['training set', 'validation set'], loc='upper right')
plt.show()
load data start! The epoch10 mkv is processing loading data filished! 2700/2700 [==============================] - 0s Test loss is:4.479728813877812
In [6]:
### <5>保存model
import os
# model文件和json文件存储路径
model_saved_path = os.path.join('./models/', "nvidia_refined_model.h5")
json_saved_path = os.path.join('./models/', "nvidia_refined_model.json")
# 存储json
json_model = model.to_json()
with open(json_saved_path, "w") as json_file:
json_file.write(json_model)
# 存储model
model.save(model_saved_path)
(3) Refined Model + 转向角度数据增加¶
In [1]:
### <1>用generator 生成NVIDIA refined model训练数据
import csv
import preprocess_data
import random
from sklearn.model_selection import train_test_split
#修改读取samples的function'read_csv',增加转向角度小于-5或大于5的sample
#原samples大小为24300，现增加为40000
def read_csv_add():
"""
# Reading the csv file and store in the list
"""
samples_original = []
samples_steering = []
with open('./train_images/train_images.csv') as csvfile:
reader = csv.reader(csvfile)
for line in reader:
samples_original.append(line)
samples_original = samples_original[1:]
# 提取转向角度小于-5或大于5的sample
for sample in samples_original:
if (float(sample[0]) < -5.0) or (float(sample[0]) > 5.0):
samples_steering.append(sample)
#增加转向幅度大的sample
while True:
if len(samples_original) < 40000:
#print(len(samples_steering))
index = random.choice(range(len(samples_steering)))
samples_original.append(samples_steering[index])
else:
break
return samples_original
#载入训练数据
samples = read_csv_add()
print(len(samples))
40000
In [2]:
# 划分训练集和验证集
train_samples, validation_samples = train_test_split(samples, test_size=0.2)
train_generator = preprocess_data.generator(train_samples, batch_size=128)
validation_generator = preprocess_data.generator(validation_samples, batch_size=128)
In [3]:
### <2>建立NVIDIA refined model
from keras.models import Sequential
from keras.layers import Convolution2D, Dropout, Flatten, Dense, Lambda, BatchNormalization, MaxPooling2D
def Nvidia_refined_model():
"""
Nvidia refined model
"""
print("Nvidia-------------")
model = Sequential()
model.add(Convolution2D(24,5,5,subsample=(2,2),activation='elu',border_mode='same',init='he_normal',input_shape=(80,80,3)))
model.add(BatchNormalization())
model.add(MaxPooling2D())
model.add(Convolution2D(36,5,5,subsample=(2,2),activation='elu',border_mode='same',init='he_normal'))
model.add(BatchNormalization())
model.add(MaxPooling2D())
model.add(Convolution2D(48,5,5,subsample=(2,2),activation='elu',border_mode='same',init='he_normal'))
model.add(BatchNormalization())
model.add(MaxPooling2D())
model.add(Convolution2D(64,3,3,activation='elu',border_mode='same',init='he_normal'))
model.add(BatchNormalization())
model.add(Convolution2D(64,3,3,activation='elu',border_mode='same',init='he_normal'))
model.add(BatchNormalization())
model.add(Flatten())
model.add(Dense(1164, activation='elu',init='he_normal'))
model.add(Dropout(0.5))
model.add(Dense(100, activation='elu',init='he_normal'))
model.add(Dropout(0.3))
model.add(Dense(50, activation='elu',init='he_normal'))
model.add(Dropout(0.2))
model.add(Dense(10, activation='elu',init='he_normal'))
model.add(Dropout(0.1))
model.add(Dense(1))
return model
Using TensorFlow backend.
In [4]:
Nvidia_refined_model().summary()
Nvidia------------- ____________________________________________________________________________________________________ Layer (type) Output Shape Param # Connected to ==================================================================================================== convolution2d_1 (Convolution2D) (None, 40, 40, 24) 1824 convolution2d_input_1[0][0] ____________________________________________________________________________________________________ batchnormalization_1 (BatchNorma (None, 40, 40, 24) 96 convolution2d_1[0][0] ____________________________________________________________________________________________________ maxpooling2d_1 (MaxPooling2D) (None, 20, 20, 24) 0 batchnormalization_1[0][0] ____________________________________________________________________________________________________ convolution2d_2 (Convolution2D) (None, 10, 10, 36) 21636 maxpooling2d_1[0][0] ____________________________________________________________________________________________________ batchnormalization_2 (BatchNorma (None, 10, 10, 36) 144 convolution2d_2[0][0] ____________________________________________________________________________________________________ maxpooling2d_2 (MaxPooling2D) (None, 5, 5, 36) 0 batchnormalization_2[0][0] ____________________________________________________________________________________________________ convolution2d_3 (Convolution2D) (None, 3, 3, 48) 43248 maxpooling2d_2[0][0] ____________________________________________________________________________________________________ batchnormalization_3 (BatchNorma (None, 3, 3, 48) 192 convolution2d_3[0][0] ____________________________________________________________________________________________________ maxpooling2d_3 (MaxPooling2D) (None, 1, 1, 48) 0 batchnormalization_3[0][0] ____________________________________________________________________________________________________ convolution2d_4 (Convolution2D) (None, 1, 1, 64) 27712 maxpooling2d_3[0][0] ____________________________________________________________________________________________________ batchnormalization_4 (BatchNorma (None, 1, 1, 64) 256 convolution2d_4[0][0] ____________________________________________________________________________________________________ convolution2d_5 (Convolution2D) (None, 1, 1, 64) 36928 batchnormalization_4[0][0] ____________________________________________________________________________________________________ batchnormalization_5 (BatchNorma (None, 1, 1, 64) 256 convolution2d_5[0][0] ____________________________________________________________________________________________________ flatten_1 (Flatten) (None, 64) 0 batchnormalization_5[0][0] ____________________________________________________________________________________________________ dense_1 (Dense) (None, 1164) 75660 flatten_1[0][0] ____________________________________________________________________________________________________ dropout_1 (Dropout) (None, 1164) 0 dense_1[0][0] ____________________________________________________________________________________________________ dense_2 (Dense) (None, 100) 116500 dropout_1[0][0] ____________________________________________________________________________________________________ dropout_2 (Dropout) (None, 100) 0 dense_2[0][0] ____________________________________________________________________________________________________ dense_3 (Dense) (None, 50) 5050 dropout_2[0][0] ____________________________________________________________________________________________________ dropout_3 (Dropout) (None, 50) 0 dense_3[0][0] ____________________________________________________________________________________________________ dense_4 (Dense) (None, 10) 510 dropout_3[0][0] ____________________________________________________________________________________________________ dropout_4 (Dropout) (None, 10) 0 dense_4[0][0] ____________________________________________________________________________________________________ dense_5 (Dense) (None, 1) 11 dropout_4[0][0] ==================================================================================================== Total params: 330,023 Trainable params: 329,551 Non-trainable params: 472 ____________________________________________________________________________________________________
In [5]:
### <3>NVIDIA refined model模型训练
model = Nvidia_refined_model()
model.compile(loss='mse', optimizer='adam')
history_object = model.fit_generator(train_generator, samples_per_epoch=
len(train_samples)*3, validation_data=validation_generator,
nb_val_samples=len(validation_samples)*3,nb_epoch=10)
Nvidia------------- Epoch 1/10 96000/96000 [==============================] - 265s - loss: 12.0565 - val_loss: 24.0982 Epoch 2/10 96000/96000 [==============================] - 61s - loss: 4.1674 - val_loss: 3.9496 Epoch 3/10 96000/96000 [==============================] - 61s - loss: 3.1119 - val_loss: 1.3425 Epoch 4/10 96000/96000 [==============================] - 60s - loss: 2.6613 - val_loss: 1.1643 Epoch 5/10 96000/96000 [==============================] - 62s - loss: 2.3794 - val_loss: 0.8630 Epoch 6/10 96000/96000 [==============================] - 61s - loss: 2.1848 - val_loss: 0.7885 Epoch 7/10 96000/96000 [==============================] - 62s - loss: 2.0637 - val_loss: 0.6422 Epoch 8/10 96000/96000 [==============================] - 62s - loss: 1.9930 - val_loss: 0.8113 Epoch 9/10 96000/96000 [==============================] - 63s - loss: 1.8898 - val_loss: 0.5942 Epoch 10/10 96000/96000 [==============================] - 63s - loss: 1.8166 - val_loss: 0.5717
In [6]:
### <4>分析结果
import preprocess_data
import matplotlib.pyplot as plt
#读取测试测试文件
test_imgs, test_wheels = preprocess_data.load_data('test')
test_imgs = preprocess_data.nomorlize_image(test_imgs)
# 测试模型在测试文件上表现
test_loss= model.evaluate(test_imgs, test_wheels, batch_size=128)
print('\n Test loss is:{}'.format(test_loss))
# 查看训练集和验证集loss
plt.plot(history_object.history['loss'])
plt.plot(history_object.history['val_loss'])
plt.title('Nvidia model mean squared error loss')
plt.ylabel('mean squared error loss')
plt.xlabel('epoch')
plt.legend(['training set', 'validation set'], loc='upper right')
plt.show()
load data start! The epoch10 mkv is processing loading data filished! 2700/2700 [==============================] - 0s Test loss is:3.2823049792536985
In [7]:
### <5>保存model
import os
# model文件和json文件存储路径
model_saved_path = os.path.join('./models/', "nvidia_ra_model.h5")
json_saved_path = os.path.join('./models/', "nvidia_ra_model.json")
# 存储json
json_model = model.to_json()
with open(json_saved_path, "w") as json_file:
json_file.write(json_model)
# 存储model
model.save(model_saved_path)
分析：增加转向幅度大的sample后，训练数据达到96000，模型在验证集和测试集上有所提升。说明增加后对模型是有帮助的，同时还需要结合时序图来分析对比预测转向和人工转向的差别。
2.VGG16 + Nvidia model
(1)notop + noblocks¶
In [1]:
### <1>.用generator生成VGG16 + Nvidia model 训练数据
import preprocess_data
from sklearn.model_selection import train_test_split
#载入训练数据
samples = preprocess_data.read_csv('./train_images/train_images.csv')
train_samples, validation_samples = train_test_split(samples, test_size=0.2)
train_generator = preprocess_data.generator(train_samples, batch_size=64)
validation_generator = preprocess_data.generator(validation_samples, batch_size=64)
In [2]:
### <2>VGG16 + Nvidia model
from keras.models import Sequential
from keras.layers import Convolution2D, Dropout, Flatten, Dense, Lambda, Input, GlobalAveragePooling2D
from keras.models import Model
from keras.applications.vgg16 import VGG16
def Vgg16_first_model():
"""
Vgg16 model
"""
print("Vgg16---------------")
input_shape = (80,80,3)
input_tensor = Input(shape=input_shape)
base_model = VGG16(input_tensor=input_tensor, weights='imagenet', include_top=False)
x = base_model.output
x = GlobalAveragePooling2D()(x)
x = Dense(512, activation='elu')(x)
x = Dropout(0.5)(x)
x = Dense(256, activation='elu')(x)
x = Dropout(0.3)(x)
x = Dense(64, activation='elu')(x)
x = Dropout(0.1)(x)
predictions = Dense(1, init='zero')(x)
# the model we will train
model = Model(input=base_model.input, output=predictions)
return model
Using TensorFlow backend.
In [3]:
Vgg16_first_model().summary()
Vgg16--------------- ____________________________________________________________________________________________________ Layer (type) Output Shape Param # Connected to ==================================================================================================== input_1 (InputLayer) (None, 80, 80, 3) 0 ____________________________________________________________________________________________________ block1_conv1 (Convolution2D) (None, 80, 80, 64) 1792 input_1[0][0] ____________________________________________________________________________________________________ block1_conv2 (Convolution2D) (None, 80, 80, 64) 36928 block1_conv1[0][0] ____________________________________________________________________________________________________ block1_pool (MaxPooling2D) (None, 40, 40, 64) 0 block1_conv2[0][0] ____________________________________________________________________________________________________ block2_conv1 (Convolution2D) (None, 40, 40, 128) 73856 block1_pool[0][0] ____________________________________________________________________________________________________ block2_conv2 (Convolution2D) (None, 40, 40, 128) 147584 block2_conv1[0][0] ____________________________________________________________________________________________________ block2_pool (MaxPooling2D) (None, 20, 20, 128) 0 block2_conv2[0][0] ____________________________________________________________________________________________________ block3_conv1 (Convolution2D) (None, 20, 20, 256) 295168 block2_pool[0][0] ____________________________________________________________________________________________________ block3_conv2 (Convolution2D) (None, 20, 20, 256) 590080 block3_conv1[0][0] ____________________________________________________________________________________________________ block3_conv3 (Convolution2D) (None, 20, 20, 256) 590080 block3_conv2[0][0] ____________________________________________________________________________________________________ block3_pool (MaxPooling2D) (None, 10, 10, 256) 0 block3_conv3[0][0] ____________________________________________________________________________________________________ block4_conv1 (Convolution2D) (None, 10, 10, 512) 1180160 block3_pool[0][0] ____________________________________________________________________________________________________ block4_conv2 (Convolution2D) (None, 10, 10, 512) 2359808 block4_conv1[0][0] ____________________________________________________________________________________________________ block4_conv3 (Convolution2D) (None, 10, 10, 512) 2359808 block4_conv2[0][0] ____________________________________________________________________________________________________ block4_pool (MaxPooling2D) (None, 5, 5, 512) 0 block4_conv3[0][0] ____________________________________________________________________________________________________ block5_conv1 (Convolution2D) (None, 5, 5, 512) 2359808 block4_pool[0][0] ____________________________________________________________________________________________________ block5_conv2 (Convolution2D) (None, 5, 5, 512) 2359808 block5_conv1[0][0] ____________________________________________________________________________________________________ block5_conv3 (Convolution2D) (None, 5, 5, 512) 2359808 block5_conv2[0][0] ____________________________________________________________________________________________________ block5_pool (MaxPooling2D) (None, 2, 2, 512) 0 block5_conv3[0][0] ____________________________________________________________________________________________________ globalaveragepooling2d_1 (Global (None, 512) 0 block5_pool[0][0] ____________________________________________________________________________________________________ dense_1 (Dense) (None, 512) 262656 globalaveragepooling2d_1[0][0] ____________________________________________________________________________________________________ dropout_1 (Dropout) (None, 512) 0 dense_1[0][0] ____________________________________________________________________________________________________ dense_2 (Dense) (None, 256) 131328 dropout_1[0][0] ____________________________________________________________________________________________________ dropout_2 (Dropout) (None, 256) 0 dense_2[0][0] ____________________________________________________________________________________________________ dense_3 (Dense) (None, 64) 16448 dropout_2[0][0] ____________________________________________________________________________________________________ dropout_3 (Dropout) (None, 64) 0 dense_3[0][0] ____________________________________________________________________________________________________ dense_4 (Dense) (None, 1) 65 dropout_3[0][0] ==================================================================================================== Total params: 15,125,185 Trainable params: 15,125,185 Non-trainable params: 0 ____________________________________________________________________________________________________
In [6]:
### <3>VGG16 + Nvidia模型训练
from keras.optimizers import Adam
model = Vgg16_first_model()
# 不选择blocks训练
for layer in model.layers[:19]:
layer.trainable = False
for layer in model.layers[19:]:
layer.trainable = True
# Default parameters follow those provided in the original paper.
opt = Adam(lr=0.001, beta_1=0.9, beta_2=0.999, epsilon=1e-08, decay=0.6)
model.compile(optimizer=opt, loss='mse')
history_object = model.fit_generator(train_generator, samples_per_epoch=
len(train_samples)*3, validation_data=validation_generator,
nb_val_samples=len(validation_samples)*3,nb_epoch=5)
Vgg16--------------- Epoch 1/5 58320/58320 [==============================] - 109s - loss: 21.0579 - val_loss: 21.4298 Epoch 2/5 58320/58320 [==============================] - 108s - loss: 21.0543 - val_loss: 21.4272 Epoch 3/5 58320/58320 [==============================] - 108s - loss: 21.0524 - val_loss: 21.4257 Epoch 4/5 58320/58320 [==============================] - 108s - loss: 21.0525 - val_loss: 21.4244 Epoch 5/5 58320/58320 [==============================] - 108s - loss: 21.0515 - val_loss: 21.4235
分析：可以看出，不加入block进行训练，原始参数并不能很好提取图像的特征，导致模型无法很好的收敛。所以需要加入卷积层进行训练。
(2)notop + top2 bolocks¶
In [2]:
### <1>.用generator生成VGG16 + Nvidia model 训练数据
import preprocess_data
from sklearn.model_selection import train_test_split
#载入训练数据
samples = preprocess_data.read_csv('./train_images/train_images.csv')
train_samples, validation_samples = train_test_split(samples, test_size=0.2)
train_generator = preprocess_data.generator(train_samples, batch_size=64)
validation_generator = preprocess_data.generator(validation_samples, batch_size=64)
In [3]:
### <2>VGG16 + Nvidia model
from keras.models import Sequential
from keras.layers import Convolution2D, Dropout, Flatten, Dense, Lambda, Input, GlobalAveragePooling2D
from keras.models import Model
from keras.applications.vgg16 import VGG16
def Vgg16_transfer_model():
"""
Vgg16 model
"""
print("Vgg16---------------")
input_shape = (80,80,3)
input_tensor = Input(shape=input_shape)
base_model = VGG16(input_tensor=input_tensor, weights='imagenet', include_top=False)
x = base_model.output
x = GlobalAveragePooling2D()(x)
x = Dense(512, activation='elu')(x)
x = Dropout(0.5)(x)
x = Dense(256, activation='elu')(x)
x = Dropout(0.3)(x)
x = Dense(64, activation='elu')(x)
x = Dropout(0.1)(x)
predictions = Dense(1, init='zero')(x)
# the model we will train
model = Model(input=base_model.input, output=predictions)
return model
Using TensorFlow backend.
In [4]:
Vgg16_transfer_model().summary()
Vgg16--------------- ____________________________________________________________________________________________________ Layer (type) Output Shape Param # Connected to ==================================================================================================== input_1 (InputLayer) (None, 80, 80, 3) 0 ____________________________________________________________________________________________________ block1_conv1 (Convolution2D) (None, 80, 80, 64) 1792 input_1[0][0] ____________________________________________________________________________________________________ block1_conv2 (Convolution2D) (None, 80, 80, 64) 36928 block1_conv1[0][0] ____________________________________________________________________________________________________ block1_pool (MaxPooling2D) (None, 40, 40, 64) 0 block1_conv2[0][0] ____________________________________________________________________________________________________ block2_conv1 (Convolution2D) (None, 40, 40, 128) 73856 block1_pool[0][0] ____________________________________________________________________________________________________ block2_conv2 (Convolution2D) (None, 40, 40, 128) 147584 block2_conv1[0][0] ____________________________________________________________________________________________________ block2_pool (MaxPooling2D) (None, 20, 20, 128) 0 block2_conv2[0][0] ____________________________________________________________________________________________________ block3_conv1 (Convolution2D) (None, 20, 20, 256) 295168 block2_pool[0][0] ____________________________________________________________________________________________________ block3_conv2 (Convolution2D) (None, 20, 20, 256) 590080 block3_conv1[0][0] ____________________________________________________________________________________________________ block3_conv3 (Convolution2D) (None, 20, 20, 256) 590080 block3_conv2[0][0] ____________________________________________________________________________________________________ block3_pool (MaxPooling2D) (None, 10, 10, 256) 0 block3_conv3[0][0] ____________________________________________________________________________________________________ block4_conv1 (Convolution2D) (None, 10, 10, 512) 1180160 block3_pool[0][0] ____________________________________________________________________________________________________ block4_conv2 (Convolution2D) (None, 10, 10, 512) 2359808 block4_conv1[0][0] ____________________________________________________________________________________________________ block4_conv3 (Convolution2D) (None, 10, 10, 512) 2359808 block4_conv2[0][0] ____________________________________________________________________________________________________ block4_pool (MaxPooling2D) (None, 5, 5, 512) 0 block4_conv3[0][0] ____________________________________________________________________________________________________ block5_conv1 (Convolution2D) (None, 5, 5, 512) 2359808 block4_pool[0][0] ____________________________________________________________________________________________________ block5_conv2 (Convolution2D) (None, 5, 5, 512) 2359808 block5_conv1[0][0] ____________________________________________________________________________________________________ block5_conv3 (Convolution2D) (None, 5, 5, 512) 2359808 block5_conv2[0][0] ____________________________________________________________________________________________________ block5_pool (MaxPooling2D) (None, 2, 2, 512) 0 block5_conv3[0][0] ____________________________________________________________________________________________________ globalaveragepooling2d_1 (Global (None, 512) 0 block5_pool[0][0] ____________________________________________________________________________________________________ dense_1 (Dense) (None, 512) 262656 globalaveragepooling2d_1[0][0] ____________________________________________________________________________________________________ dropout_1 (Dropout) (None, 512) 0 dense_1[0][0] ____________________________________________________________________________________________________ dense_2 (Dense) (None, 256) 131328 dropout_1[0][0] ____________________________________________________________________________________________________ dropout_2 (Dropout) (None, 256) 0 dense_2[0][0] ____________________________________________________________________________________________________ dense_3 (Dense) (None, 64) 16448 dropout_2[0][0] ____________________________________________________________________________________________________ dropout_3 (Dropout) (None, 64) 0 dense_3[0][0] ____________________________________________________________________________________________________ dense_4 (Dense) (None, 1) 65 dropout_3[0][0] ==================================================================================================== Total params: 15,125,185 Trainable params: 15,125,185 Non-trainable params: 0 ____________________________________________________________________________________________________
In [4]:
### <3>VGG16 + Nvidia模型训练
from keras.optimizers import Adam
model = Vgg16_transfer_model()
# 选择top2 blocks训练
for layer in model.layers[:11]:
layer.trainable = False
for layer in model.layers[11:]:
layer.trainable = True
# Default parameters follow those provided in the original paper.
opt = Adam(lr=0.001, beta_1=0.9, beta_2=0.999, epsilon=1e-08, decay=0.6)
model.compile(optimizer=opt, loss='mse')
history_object = model.fit_generator(train_generator, samples_per_epoch=
len(train_samples)*3, validation_data=validation_generator,
nb_val_samples=len(validation_samples)*3,nb_epoch=10)
Vgg16--------------- Epoch 1/10 58320/58320 [==============================] - 172s - loss: 14.5413 - val_loss: 10.7658 Epoch 2/10 58320/58320 [==============================] - 165s - loss: 10.0365 - val_loss: 8.2215 Epoch 3/10 58320/58320 [==============================] - 165s - loss: 8.2899 - val_loss: 7.0026 Epoch 4/10 58320/58320 [==============================] - 165s - loss: 7.2961 - val_loss: 6.2059 Epoch 5/10 58320/58320 [==============================] - 166s - loss: 6.6702 - val_loss: 5.7412 Epoch 6/10 58320/58320 [==============================] - 166s - loss: 6.2682 - val_loss: 5.3753 Epoch 7/10 58320/58320 [==============================] - 166s - loss: 5.9600 - val_loss: 5.1232 Epoch 8/10 58320/58320 [==============================] - 165s - loss: 5.6579 - val_loss: 4.8521 Epoch 9/10 58320/58320 [==============================] - 165s - loss: 5.4689 - val_loss: 4.6580 Epoch 10/10 58320/58320 [==============================] - 165s - loss: 5.3162 - val_loss: 4.5624
In [5]:
### <4>分析结果
import preprocess_data
import matplotlib.pyplot as plt
#读取测试测试文件
test_imgs, test_wheels = preprocess_data.load_data('test')
test_imgs = preprocess_data.nomorlize_image(test_imgs)
# 测试模型在测试文件上表现
test_loss= model.evaluate(test_imgs, test_wheels, batch_size=128)
print('\n Test loss is:{}'.format(test_loss))
# 查看训练集和验证集loss
plt.plot(history_object.history['loss'])
plt.plot(history_object.history['val_loss'])
plt.title('VGG16 + Nvidia mean squared error loss')
plt.ylabel('mean squared error loss')
plt.xlabel('epoch')
plt.legend(['training set', 'validation set'], loc='upper right')
plt.show()
load data start! The epoch10 mkv is processing loading data filished! 2700/2700 [==============================] - 6s Test loss is:2.782660860132288
In [6]:
### <5>保存model
import os
# model文件和json文件存储路径
model_saved_path = os.path.join('./models/', "vgg16_model.h5")
json_saved_path = os.path.join('./models/', "vgg16_model.json")
# 存储json
json_model = model.to_json()
with open(json_saved_path, "w") as json_file:
json_file.write(json_model)
# 存储model
model.save(model_saved_path)
(3)notop + top2 bolocks + 转向角度数据增加¶
In [1]:
### <1>用generator 生成Vgg16 + Nvidia 训练数据
import csv
import preprocess_data
import random
from sklearn.model_selection import train_test_split
#修改读取samples的function'read_csv',增加转向角度小于-5或大于5的sample
#原samples大小为24300，现增加为40000
def read_csv_add():
"""
# Reading the csv file and store in the list
"""
samples_original = []
samples_steering = []
with open('./train_images/train_images.csv') as csvfile:
reader = csv.reader(csvfile)
for line in reader:
samples_original.append(line)
samples_original = samples_original[1:]
# 提取转向角度小于-5或大于5的sample
for sample in samples_original:
if (float(sample[0]) < -5.0) or (float(sample[0]) > 5.0):
samples_steering.append(sample)
#增加转向幅度大的sample
while True:
if len(samples_original) < 40000:
#print(len(samples_steering))
index = random.choice(range(len(samples_steering)))
samples_original.append(samples_steering[index])
else:
break
return samples_original
#载入训练数据
samples = read_csv_add()
print(len(samples))
40000
In [2]:
# 划分训练集和验证集
train_samples, validation_samples = train_test_split(samples, test_size=0.2)
train_generator = preprocess_data.generator(train_samples, batch_size=128)
validation_generator = preprocess_data.generator(validation_samples, batch_size=128)
In [3]:
### <2>VGG16 + Nvidia model
from keras.models import Sequential
from keras.layers import Convolution2D, Dropout, Flatten, Dense, Lambda, Input, GlobalAveragePooling2D
from keras.models import Model
from keras.applications.vgg16 import VGG16
def Vgg16_transfer_model():
"""
Vgg16 model
"""
print("Vgg16---------------")
input_shape = (80,80,3)
input_tensor = Input(shape=input_shape)
base_model = VGG16(input_tensor=input_tensor, weights='imagenet', include_top=False)
x = base_model.output
x = GlobalAveragePooling2D()(x)
x = Dense(512, activation='elu')(x)
x = Dropout(0.5)(x)
x = Dense(256, activation='elu')(x)
x = Dropout(0.3)(x)
x = Dense(64, activation='elu')(x)
x = Dropout(0.1)(x)
predictions = Dense(1, init='zero')(x)
# the model we will train
model = Model(input=base_model.input, output=predictions)
return model
Using TensorFlow backend.
In [4]:
Vgg16_transfer_model().summary()
Vgg16--------------- ____________________________________________________________________________________________________ Layer (type) Output Shape Param # Connected to ==================================================================================================== input_1 (InputLayer) (None, 80, 80, 3) 0 ____________________________________________________________________________________________________ block1_conv1 (Convolution2D) (None, 80, 80, 64) 1792 input_1[0][0] ____________________________________________________________________________________________________ block1_conv2 (Convolution2D) (None, 80, 80, 64) 36928 block1_conv1[0][0] ____________________________________________________________________________________________________ block1_pool (MaxPooling2D) (None, 40, 40, 64) 0 block1_conv2[0][0] ____________________________________________________________________________________________________ block2_conv1 (Convolution2D) (None, 40, 40, 128) 73856 block1_pool[0][0] ____________________________________________________________________________________________________ block2_conv2 (Convolution2D) (None, 40, 40, 128) 147584 block2_conv1[0][0] ____________________________________________________________________________________________________ block2_pool (MaxPooling2D) (None, 20, 20, 128) 0 block2_conv2[0][0] ____________________________________________________________________________________________________ block3_conv1 (Convolution2D) (None, 20, 20, 256) 295168 block2_pool[0][0] ____________________________________________________________________________________________________ block3_conv2 (Convolution2D) (None, 20, 20, 256) 590080 block3_conv1[0][0] ____________________________________________________________________________________________________ block3_conv3 (Convolution2D) (None, 20, 20, 256) 590080 block3_conv2[0][0] ____________________________________________________________________________________________________ block3_pool (MaxPooling2D) (None, 10, 10, 256) 0 block3_conv3[0][0] ____________________________________________________________________________________________________ block4_conv1 (Convolution2D) (None, 10, 10, 512) 1180160 block3_pool[0][0] ____________________________________________________________________________________________________ block4_conv2 (Convolution2D) (None, 10, 10, 512) 2359808 block4_conv1[0][0] ____________________________________________________________________________________________________ block4_conv3 (Convolution2D) (None, 10, 10, 512) 2359808 block4_conv2[0][0] ____________________________________________________________________________________________________ block4_pool (MaxPooling2D) (None, 5, 5, 512) 0 block4_conv3[0][0] ____________________________________________________________________________________________________ block5_conv1 (Convolution2D) (None, 5, 5, 512) 2359808 block4_pool[0][0] ____________________________________________________________________________________________________ block5_conv2 (Convolution2D) (None, 5, 5, 512) 2359808 block5_conv1[0][0] ____________________________________________________________________________________________________ block5_conv3 (Convolution2D) (None, 5, 5, 512) 2359808 block5_conv2[0][0] ____________________________________________________________________________________________________ block5_pool (MaxPooling2D) (None, 2, 2, 512) 0 block5_conv3[0][0] ____________________________________________________________________________________________________ globalaveragepooling2d_1 (Global (None, 512) 0 block5_pool[0][0] ____________________________________________________________________________________________________ dense_1 (Dense) (None, 512) 262656 globalaveragepooling2d_1[0][0] ____________________________________________________________________________________________________ dropout_1 (Dropout) (None, 512) 0 dense_1[0][0] ____________________________________________________________________________________________________ dense_2 (Dense) (None, 256) 131328 dropout_1[0][0] ____________________________________________________________________________________________________ dropout_2 (Dropout) (None, 256) 0 dense_2[0][0] ____________________________________________________________________________________________________ dense_3 (Dense) (None, 64) 16448 dropout_2[0][0] ____________________________________________________________________________________________________ dropout_3 (Dropout) (None, 64) 0 dense_3[0][0] ____________________________________________________________________________________________________ dense_4 (Dense) (None, 1) 65 dropout_3[0][0] ==================================================================================================== Total params: 15,125,185 Trainable params: 15,125,185 Non-trainable params: 0 ____________________________________________________________________________________________________
In [5]:
### <3>VGG16 + Nvidia模型训练
from keras.optimizers import Adam
model = Vgg16_transfer_model()
# 选择top2 blocks训练
for layer in model.layers[:11]:
layer.trainable = False
for layer in model.layers[11:]:
layer.trainable = True
# Default parameters follow those provided in the original paper.
opt = Adam(lr=0.001, beta_1=0.9, beta_2=0.999, epsilon=1e-08, decay=0.6)
model.compile(optimizer=opt, loss='mse')
history_object = model.fit_generator(train_generator, samples_per_epoch=
len(train_samples)*3, validation_data=validation_generator,
nb_val_samples=len(validation_samples)*3,nb_epoch=10)
Vgg16--------------- Epoch 1/10 96000/96000 [==============================] - 262s - loss: 19.4276 - val_loss: 8.1972 Epoch 2/10 96000/96000 [==============================] - 257s - loss: 8.1509 - val_loss: 5.4931 Epoch 3/10 96000/96000 [==============================] - 257s - loss: 6.5458 - val_loss: 4.6628 Epoch 4/10 96000/96000 [==============================] - 257s - loss: 5.7366 - val_loss: 4.1017 Epoch 5/10 96000/96000 [==============================] - 258s - loss: 5.3208 - val_loss: 3.7532 Epoch 6/10 96000/96000 [==============================] - 258s - loss: 4.9637 - val_loss: 3.5392 Epoch 7/10 96000/96000 [==============================] - 257s - loss: 4.7431 - val_loss: 3.3827 Epoch 8/10 96000/96000 [==============================] - 258s - loss: 4.5388 - val_loss: 3.2475 Epoch 9/10 96000/96000 [==============================] - 258s - loss: 4.3775 - val_loss: 3.1380 Epoch 10/10 96000/96000 [==============================] - 258s - loss: 4.2552 - val_loss: 2.9988
In [6]:
### <4>分析结果
import preprocess_data
import matplotlib.pyplot as plt
#读取测试测试文件
test_imgs, test_wheels = preprocess_data.load_data('test')
test_imgs = preprocess_data.nomorlize_image(test_imgs)
# 测试模型在测试文件上表现
test_loss= model.evaluate(test_imgs, test_wheels, batch_size=128)
print('\n Test loss is:{}'.format(test_loss))
# 查看训练集和验证集loss
plt.plot(history_object.history['loss'])
plt.plot(history_object.history['val_loss'])
plt.title('VGG16 + Nvidia mean squared error loss')
plt.ylabel('mean squared error loss')
plt.xlabel('epoch')
plt.legend(['training set', 'validation set'], loc='upper right')
plt.show()
load data start! The epoch10 mkv is processing loading data filished! 2700/2700 [==============================] - 5s Test loss is:3.54451801971153
In [7]:
### <5>保存model
import os
# model文件和json文件存储路径
model_saved_path = os.path.join('./models/', "vgg16_add_model.h5")
json_saved_path = os.path.join('./models/', "vgg16_add_model.json")
# 存储json
json_model = model.to_json()
with open(json_saved_path, "w") as json_file:
json_file.write(json_model)
# 存储model
model.save(model_saved_path)
分析：同样，增加转向幅度大的sample后，训练数据达到96000，模型在验证集上有所提升，但测试集变化不大。再次说明数据增加后对模型帮助是有限的，同时还需要结合时序图来分析对比预测转向和人工转向的差别。
3.CNN+RNN seq2seq model （选做）
说明：
模型建立来源于：UDACITY 的开源挑战项目 9 "Teaching a Machine to Steer a Car"，其中第一名"Team Komanda"建立的seq2seq模型。
参考实现代码：https://github.com/udacity/self-driving-car/tree/master/steering-models/community-models/komanda 的 solution-komanda.ipynb文件
作为选做项目，按照原代码处理数据方式，并没有进行数据增加处理技术，直接使用原图片进行模型训练
In [1]:
### <1>导入使用的库文件
import tensorflow as tf
import numpy as np
import pandas as pd
import os
import csv
slim = tf.contrib.slim
In [2]:
### <2>定义相关宏参数
SEQ_LEN = 10
BATCH_SIZE = 4
LEFT_CONTEXT = 5
# 这些是输入图片的维度
HEIGHT = 80
WIDTH = 80
CHANNELS = 3 # RGB
# LSTM 保存模型 'state'(状态的参数)
RNN_SIZE = 32
RNN_PROJ = 32
# 要预测的是转向角度信息
CSV_HEADER = "timestamp,index,angle".split(",")
OUTPUTS = [CSV_HEADER[2]]
OUTPUT_DIM = len(OUTPUTS)
In [3]:
### <3>输入输出的格式定义
### 通过读取csv文件生成形如 BATCH_SIZE x SEQ_LEN 矩阵的samples
### 然后使用'BatchGenerator'生成器每个sequence 向前看 LEFT_CONTEXT 的图片，
### 通过生成器生成形如 [BATCH_SIZE, LEFT_CONTEXT + SEQ_LEN, HEIGHT, WIDTH, CHANNELS] 矩阵的训练数据
import cv2
class BatchGenerator(object):
def __init__(self, sequence, seq_len, batch_size):
self.sequence = sequence
self.seq_len = seq_len
self.batch_size = batch_size
chunk_size = 1 + (len(sequence) - 1) // batch_size
self.indices = [(i*chunk_size) % len(sequence) for i in range(batch_size)]
def next(self):
while True:
output = []
for i in range(self.batch_size):
idx = self.indices[i]
left_pad = self.sequence[idx - LEFT_CONTEXT:idx]
if len(left_pad) < LEFT_CONTEXT:
left_pad = [self.sequence[0]] * (LEFT_CONTEXT - len(left_pad)) + left_pad
assert len(left_pad) == LEFT_CONTEXT
leftover = len(self.sequence) - idx
if leftover >= self.seq_len:
result = self.sequence[idx:idx + self.seq_len]
else:
result = self.sequence[idx:] + self.sequence[:self.seq_len - leftover]
assert len(result) == self.seq_len
#print(self.seq_len)
#print(result)
#print(left_pad)
self.indices[i] = (idx + self.seq_len) % len(self.sequence)
images, targets = zip(*result)
images_left_pad, _ = zip(*left_pad)
output.append((np.stack(images_left_pad + images), np.stack(targets)))
#print(output)
#output = [zip(*output)]
#output = zip(*output)
#output = zip(*output[0])
output_zip = tuple(zip(*output))
#print(np.hstack(list(output_zip)[0]))
#print(np.hstack(list(output_zip)[1]))
output_0 = np.hstack(list(output_zip)[0])# batch_size x (LEFT_CONTEXT + seq_len)
output_1 = np.hstack(list(output_zip)[1]) # batch_size x seq_len x OUTPUT_DIM
#output_0 = [cv2.imread(img) for img in output_0]
output = []
output.append(np.reshape(output_0, (self.batch_size,LEFT_CONTEXT + self.seq_len)))
output.append(np.reshape(output_1, (self.batch_size,self.seq_len,OUTPUT_DIM)))
return output
def read_csv(filename):
with open(filename, 'r') as f:
lines = [ln.strip().split(",")[0:2] for ln in f.readlines()]
lines = lines[1:]
lines = map(lambda x: (x[1], np.float32(x[0])), lines) # imagefile, outputs
return lines
def process_csv(filename, val=20):
sum_f = np.float64([0.0] * OUTPUT_DIM)
sum_sq_f = np.float64([0.0] * OUTPUT_DIM)
lines = read_csv(filename)
# leave val% for validation
train_seq = []
valid_seq = []
cnt = 0
for ln in lines:
if cnt < SEQ_LEN * BATCH_SIZE * (100 - val):
train_seq.append(ln)
sum_f += ln[1]
sum_sq_f += ln[1] * ln[1]
else:
valid_seq.append(ln)
cnt += 1
cnt %= SEQ_LEN * BATCH_SIZE * 100
mean = sum_f / len(train_seq)
var = sum_sq_f / len(train_seq) - mean * mean
std = np.sqrt(var)
print(len(train_seq), len(valid_seq))
print (mean, std) # we will need these statistics to normalize the outputs (and ground truth inputs)
return (train_seq, valid_seq), (mean, std)
In [4]:
# 读取训练和测试用的所有csv文件
(train_seq, valid_seq), (mean, std) = process_csv(filename="./train_images/train_images.csv", val=5) # concatenated interpolated.csv from rosbags
#test_seq = read_csv("./test_images/test_images.csv") # interpolated.csv for testset filled with dummy values
(test_seq, _), (test_mean, test_std) = process_csv(filename="./test_images/test_images.csv", val=0)
# 读取测试用csv文件
test_wheel = pd.read_csv('./test_images/test_images.csv')
23100 1200 [-0.06969697] [ 4.60519558] 2700 0 [-1.82666667] [ 2.12810818]
In [5]:
### <4>定义"vision module" 和"the recurrent stateful cell."
### 代码具体说明可见："solution-komanda.ipynb"文件
layer_norm = lambda x: tf.contrib.layers.layer_norm(inputs=x, center=True, scale=True, activation_fn=None, trainable=True)
def get_optimizer(loss, lrate):
optimizer = tf.train.AdamOptimizer(learning_rate=lrate)
gradvars = optimizer.compute_gradients(loss)
gradients, v = zip(*gradvars)
print([x.name for x in v])
gradients, _ = tf.clip_by_global_norm(gradients, 15.0)
return optimizer.apply_gradients(zip(gradients, v))
def apply_vision_simple(image, keep_prob, batch_size, seq_len, scope=None, reuse=None):
video = tf.reshape(image, shape=[batch_size, LEFT_CONTEXT + seq_len, HEIGHT, WIDTH, CHANNELS])
with tf.variable_scope(scope, 'Vision', [image], reuse=reuse):
net = slim.convolution(video, num_outputs=64, kernel_size=[3,5,5], stride=[1,2,2], padding="VALID")
net = tf.nn.dropout(x=net, keep_prob=keep_prob)
aux1 = slim.fully_connected(tf.reshape(net[:, -seq_len:, :, :, :], [batch_size, seq_len, -1]), 128, activation_fn=None)
net = slim.convolution(net, num_outputs=64, kernel_size=[2,5,5], stride=[1,2,2], padding="VALID")
net = tf.nn.dropout(x=net, keep_prob=keep_prob)
aux2 = slim.fully_connected(tf.reshape(net[:, -seq_len:, :, :, :], [batch_size, seq_len, -1]), 128, activation_fn=None)
net = slim.convolution(net, num_outputs=64, kernel_size=[2,5,5], stride=[1,1,1], padding="VALID")
net = tf.nn.dropout(x=net, keep_prob=keep_prob)
aux3 = slim.fully_connected(tf.reshape(net[:, -seq_len:, :, :, :], [batch_size, seq_len, -1]), 128, activation_fn=None)
net = slim.convolution(net, num_outputs=64, kernel_size=[2,5,5], stride=[1,1,1], padding="VALID")
net = tf.nn.dropout(x=net, keep_prob=keep_prob)
# at this point the tensor 'net' is of shape batch_size x seq_len x ...
aux4 = slim.fully_connected(tf.reshape(net, [batch_size, seq_len, -1]), 128, activation_fn=None)
net = slim.fully_connected(tf.reshape(net, [batch_size, seq_len, -1]), 1024, activation_fn=tf.nn.relu)
net = tf.nn.dropout(x=net, keep_prob=keep_prob)
net = slim.fully_connected(net, 512, activation_fn=tf.nn.relu)
net = tf.nn.dropout(x=net, keep_prob=keep_prob)
net = slim.fully_connected(net, 256, activation_fn=tf.nn.relu)
net = tf.nn.dropout(x=net, keep_prob=keep_prob)
net = slim.fully_connected(net, 128, activation_fn=None)
return layer_norm(tf.nn.elu(net + aux1 + aux2 + aux3 + aux4)) # aux[1-4] are residual connections (shortcuts)
class SamplingRNNCell(tf.nn.rnn_cell.RNNCell):
"""Simple sampling RNN cell."""
def __init__(self, num_outputs, use_ground_truth, internal_cell):
"""
if use_ground_truth then don't sample
"""
self._num_outputs = num_outputs
self._use_ground_truth = use_ground_truth # boolean
self._internal_cell = internal_cell # may be LSTM or GRU or anything
@property
def state_size(self):
return self._num_outputs, self._internal_cell.state_size # previous output and bottleneck state
@property
def output_size(self):
return self._num_outputs # steering angle, torque, vehicle speed
def __call__(self, inputs, state, scope=None):
(visual_feats, current_ground_truth) = inputs
prev_output, prev_state_internal = state
context = tf.concat(1, [prev_output, visual_feats])
new_output_internal, new_state_internal = internal_cell(context, prev_state_internal) # here the internal cell (e.g. LSTM) is called
new_output = tf.contrib.layers.fully_connected(
inputs=tf.concat(1, [new_output_internal, prev_output, visual_feats]),
num_outputs=self._num_outputs,
activation_fn=None,
scope="OutputProjection")
# if self._use_ground_truth == True, we pass the ground truth as the state; otherwise, we use the model's predictions
return new_output, (current_ground_truth if self._use_ground_truth else new_output, new_state_internal)
In [6]:
### <5>建立模型
### 具体说明可见："solution-komanda.ipynb"文件
graph = tf.Graph()
with graph.as_default():
# inputs
learning_rate = tf.placeholder_with_default(input=1e-4, shape=())
keep_prob = tf.placeholder_with_default(input=1.0, shape=())
aux_cost_weight = tf.placeholder_with_default(input=0.1, shape=())
inputs = tf.placeholder(shape=(BATCH_SIZE,LEFT_CONTEXT+SEQ_LEN), dtype=tf.string) # pathes to png files from the central camera
targets = tf.placeholder(shape=(BATCH_SIZE,SEQ_LEN,OUTPUT_DIM), dtype=tf.float32) # seq_len x batch_size x OUTPUT_DIM
targets_normalized = (targets - mean) / std
input_images = tf.pack([tf.image.decode_jpeg(tf.read_file(x))
for x in tf.unpack(tf.reshape(inputs, shape=[(LEFT_CONTEXT+SEQ_LEN) * BATCH_SIZE]))])
input_images = -1.0 + 2.0 * tf.cast(input_images, tf.float32) / 255.0
input_images.set_shape([(LEFT_CONTEXT+SEQ_LEN) * BATCH_SIZE, HEIGHT, WIDTH, CHANNELS])
visual_conditions_reshaped = apply_vision_simple(image=input_images, keep_prob=keep_prob,
batch_size=BATCH_SIZE, seq_len=SEQ_LEN)
visual_conditions = tf.reshape(visual_conditions_reshaped, [BATCH_SIZE, SEQ_LEN, -1])
visual_conditions = tf.nn.dropout(x=visual_conditions, keep_prob=keep_prob)
rnn_inputs_with_ground_truth = (visual_conditions, targets_normalized)
rnn_inputs_autoregressive = (visual_conditions, tf.zeros(shape=(BATCH_SIZE, SEQ_LEN, OUTPUT_DIM), dtype=tf.float32))
internal_cell = tf.nn.rnn_cell.LSTMCell(num_units=RNN_SIZE, num_proj=RNN_PROJ)
cell_with_ground_truth = SamplingRNNCell(num_outputs=OUTPUT_DIM, use_ground_truth=True, internal_cell=internal_cell)
cell_autoregressive = SamplingRNNCell(num_outputs=OUTPUT_DIM, use_ground_truth=False, internal_cell=internal_cell)
def get_initial_state(complex_state_tuple_sizes):
flat_sizes = tf.nn.rnn_cell.nest.flatten(complex_state_tuple_sizes)
init_state_flat = [tf.tile(
multiples=[BATCH_SIZE, 1],
input=tf.get_variable("controller_initial_state_%d" % i, initializer=tf.zeros_initializer, shape=([1, s]), dtype=tf.float32))
for i,s in enumerate(flat_sizes)]
init_state = tf.nn.rnn_cell.nest.pack_sequence_as(complex_state_tuple_sizes, init_state_flat)
return init_state
def deep_copy_initial_state(complex_state_tuple):
flat_state = tf.nn.rnn_cell.nest.flatten(complex_state_tuple)
flat_copy = [tf.identity(s) for s in flat_state]
deep_copy = tf.nn.rnn_cell.nest.pack_sequence_as(complex_state_tuple, flat_copy)
return deep_copy
controller_initial_state_variables = get_initial_state(cell_autoregressive.state_size)
controller_initial_state_autoregressive = deep_copy_initial_state(controller_initial_state_variables)
controller_initial_state_gt = deep_copy_initial_state(controller_initial_state_variables)
with tf.variable_scope("predictor"):
out_gt, controller_final_state_gt = tf.nn.dynamic_rnn(cell=cell_with_ground_truth, inputs=rnn_inputs_with_ground_truth,
sequence_length=[SEQ_LEN]*BATCH_SIZE, initial_state=controller_initial_state_gt, dtype=tf.float32,
swap_memory=True, time_major=False)
with tf.variable_scope("predictor", reuse=True):
out_autoregressive, controller_final_state_autoregressive = tf.nn.dynamic_rnn(cell=cell_autoregressive, inputs=rnn_inputs_autoregressive,
sequence_length=[SEQ_LEN]*BATCH_SIZE, initial_state=controller_initial_state_autoregressive, dtype=tf.float32,
swap_memory=True, time_major=False)
mse_gt = tf.reduce_mean(tf.squared_difference(out_gt, targets_normalized))
mse_autoregressive = tf.reduce_mean(tf.squared_difference(out_autoregressive, targets_normalized))
mse_autoregressive_steering = tf.reduce_mean(tf.squared_difference(out_autoregressive[:, :, 0], targets_normalized[:, :, 0]))
steering_predictions = (out_autoregressive[:, :, 0] * std[0]) + mean[0]
total_loss = mse_autoregressive_steering + aux_cost_weight * (mse_gt + mse_autoregressive)
optimizer = get_optimizer(total_loss, learning_rate)
tf.summary.scalar("MAIN_TRAIN_METRIC__mse_autoregressive_steering", mse_autoregressive_steering)
tf.summary.scalar("mse_gt", mse_gt)
tf.summary.scalar("mse_autoregressive", mse_autoregressive)
summaries = tf.merge_all_summaries()
train_writer = tf.summary.FileWriter('v3/train_summary', graph=graph)
valid_writer = tf.summary.FileWriter('v3/valid_summary', graph=graph)
saver = tf.train.Saver(write_version=tf.train.SaverDef.V2)
['Vision/Conv/weights:0', 'Vision/Conv/biases:0', 'Vision/fully_connected/weights:0', 'Vision/fully_connected/biases:0', 'Vision/Conv_1/weights:0', 'Vision/Conv_1/biases:0', 'Vision/fully_connected_1/weights:0', 'Vision/fully_connected_1/biases:0', 'Vision/Conv_2/weights:0', 'Vision/Conv_2/biases:0', 'Vision/fully_connected_2/weights:0', 'Vision/fully_connected_2/biases:0', 'Vision/Conv_3/weights:0', 'Vision/Conv_3/biases:0', 'Vision/fully_connected_3/weights:0', 'Vision/fully_connected_3/biases:0', 'Vision/fully_connected_4/weights:0', 'Vision/fully_connected_4/biases:0', 'Vision/fully_connected_5/weights:0', 'Vision/fully_connected_5/biases:0', 'Vision/fully_connected_6/weights:0', 'Vision/fully_connected_6/biases:0', 'Vision/fully_connected_7/weights:0', 'Vision/fully_connected_7/biases:0', 'Vision/LayerNorm/beta:0', 'Vision/LayerNorm/gamma:0', 'controller_initial_state_0:0', 'controller_initial_state_1:0', 'controller_initial_state_2:0', 'predictor/RNN/LSTMCell/W_0:0', 'predictor/RNN/LSTMCell/B:0', 'predictor/RNN/LSTMCell/W_P_0:0', 'predictor/RNN/OutputProjection/weights:0', 'predictor/RNN/OutputProjection/biases:0'] WARNING:tensorflow:From <ipython-input-6-4d1293e86359>:72 in <module>.: merge_all_summaries (from tensorflow.python.ops.logging_ops) is deprecated and will be removed after 2016-11-30. Instructions for updating: Please switch to tf.summary.merge_all. WARNING:tensorflow:From D:\Anaconda3\envs\carnd-term1\lib\site-packages\tensorflow\python\ops\logging_ops.py:264 in merge_all_summaries.: merge_summary (from tensorflow.python.ops.logging_ops) is deprecated and will be removed after 2016-11-30. Instructions for updating: Please switch to tf.summary.merge.
In [7]:
### <5>训练模型
### 具体说明可见："solution-komanda.ipynb"文件
gpu_options = tf.GPUOptions(per_process_gpu_memory_fraction=1.0)
checkpoint_dir = os.getcwd() + "/v3"
global_train_step = 0
global_valid_step = 0
KEEP_PROB_TRAIN = 0.25
def do_epoch(session, sequences, mode):
global global_train_step, global_valid_step
test_predictions = {}
valid_predictions = {}
batch_generator = BatchGenerator(sequence=sequences, seq_len=SEQ_LEN, batch_size=BATCH_SIZE)
total_num_steps = int(1 + (batch_generator.indices[1] - 1) // SEQ_LEN)
controller_final_state_gt_cur, controller_final_state_autoregressive_cur = None, None
acc_loss = np.float64(0.0)
for step in range(total_num_steps):
feed_inputs, feed_targets = batch_generator.next()
feed_dict = {inputs : feed_inputs, targets : feed_targets}
if controller_final_state_autoregressive_cur is not None:
feed_dict.update({controller_initial_state_autoregressive : controller_final_state_autoregressive_cur})
if controller_final_state_gt_cur is not None:
feed_dict.update({controller_final_state_gt : controller_final_state_gt_cur})
if mode == "train":
feed_dict.update({keep_prob : KEEP_PROB_TRAIN})
summary, _, loss, controller_final_state_gt_cur, controller_final_state_autoregressive_cur = \
session.run([summaries, optimizer, mse_autoregressive_steering, controller_final_state_gt, controller_final_state_autoregressive],
feed_dict = feed_dict)
train_writer.add_summary(summary, global_train_step)
global_train_step += 1
elif mode == "valid":
model_predictions, summary, loss, controller_final_state_autoregressive_cur = \
session.run([steering_predictions, summaries, mse_autoregressive_steering, controller_final_state_autoregressive],
feed_dict = feed_dict)
valid_writer.add_summary(summary, global_valid_step)
global_valid_step += 1
feed_inputs = feed_inputs[:, LEFT_CONTEXT:].flatten()
steering_targets = feed_targets[:, :, 0].flatten()
model_predictions = model_predictions.flatten()
stats = np.stack([steering_targets, model_predictions, (steering_targets - model_predictions)**2])
for i, img in enumerate(feed_inputs):
valid_predictions[img] = stats[:, i]
elif mode == "test":
model_predictions, loss, controller_final_state_autoregressive_cur = \
session.run([steering_predictions,mse_autoregressive_steering, controller_final_state_autoregressive],
feed_dict = feed_dict)
feed_inputs = feed_inputs[:, LEFT_CONTEXT:].flatten()
steering_targets = feed_targets[:, :, 0].flatten()
model_predictions = model_predictions.flatten()
stats = np.stack([steering_targets, model_predictions, (steering_targets - model_predictions)**2])
for i, img in enumerate(feed_inputs):
test_predictions[img] = stats[:, i]
if mode != "test":
acc_loss += loss
print('\r', step + 1, "/", total_num_steps,'The '+ mode + ' loss', acc_loss / (step+1))
print()
return (np.sqrt(acc_loss / total_num_steps), valid_predictions) if mode != "test" else (None, test_predictions)
NUM_EPOCHS=20
best_validation_score = None
with tf.Session(graph=graph) as session:
session.run(tf.initialize_all_variables())
print('Initialized')
ckpt = tf.train.latest_checkpoint(checkpoint_dir)
if ckpt:
print("Restoring from", ckpt)
saver.restore(sess=session, save_path=ckpt)
for epoch in range(NUM_EPOCHS):
print("Starting epoch %d" % epoch)
print("Validation:")
valid_score, valid_predictions = do_epoch(session=session, sequences=valid_seq, mode="valid")
if best_validation_score is None:
best_validation_score = valid_score
if valid_score < best_validation_score:
saver.save(session, 'v3/checkpoint-sdc-ch2')
best_validation_score = valid_score
#print( '\r', "SAVED at epoch %d" % epoch)
with open("v3/valid-predictions-epoch%d" % epoch, "w") as out:
result = np.float64(0.0)
for img, stats in valid_predictions.items():
#print(out, img, stats)
result += stats[-1]
print("Validation MSE(val_loss):", result / len(valid_predictions))
with open("v3/test-predictions-epoch%d" % epoch, "w") as out:
_, test_predictions = do_epoch(session=session, sequences=test_seq, mode="test")
#print("frame_id,steering_angle", file=out)
result = np.float64(0.0)
for img, stats in test_predictions.items():
#original_wheel = test_wheel[test_wheel['img_path']==img]['wheel']
#original_wheel_normalized = (float(original_wheel) - test_mean) / test_std
#result += np.square(float(pred) - float(original_wheel_normalized))
#img = img.replace("challenge_2/Test-final/center/", "")
#print("%s,%f" % (img, pred), file=out)
result += stats[-1]
print("Test MSE(test_loss):", result / len(test_predictions))
if epoch != NUM_EPOCHS - 1:
print("Training")
do_epoch(session=session, sequences=train_seq, mode="train")
WARNING:tensorflow:From <ipython-input-7-85165e56734f>:66 in <module>.: initialize_all_variables (from tensorflow.python.ops.variables) is deprecated and will be removed after 2017-03-02. Instructions for updating: Use `tf.global_variables_initializer` instead. Initialized Starting epoch 0 Validation: 1 / 30 The valid loss 1.22847342491 2 / 30 The valid loss 1.30617243052 3 / 30 The valid loss 1.32276403904 4 / 30 The valid loss 1.29203206301 5 / 30 The valid loss 1.21434497833 6 / 30 The valid loss 1.25092792511 7 / 30 The valid loss 1.38976485389 8 / 30 The valid loss 1.53838163614 9 / 30 The valid loss 1.7575340271 10 / 30 The valid loss 1.91051859856 11 / 30 The valid loss 1.92740091411 12 / 30 The valid loss 1.95399359862 13 / 30 The valid loss 1.97200336823 14 / 30 The valid loss 1.99875029496 15 / 30 The valid loss 1.97106359005 16 / 30 The valid loss 1.97924449295 17 / 30 The valid loss 1.99758729514 18 / 30 The valid loss 2.02590615882 19 / 30 The valid loss 2.03331466098 20 / 30 The valid loss 2.01283909082 21 / 30 The valid loss 1.96998377073 22 / 30 The valid loss 1.94439368898 23 / 30 The valid loss 1.92486905015 24 / 30 The valid loss 1.91744978229 25 / 30 The valid loss 1.88586859226 26 / 30 The valid loss 1.84927611626 27 / 30 The valid loss 1.81145428728 28 / 30 The valid loss 1.7709134881 29 / 30 The valid loss 1.72623381327 30 / 30 The valid loss 1.69248875777 Training 1 / 578 The train loss 9.84366512299 2 / 578 The train loss 8.36939239502 3 / 578 The train loss 7.6472380956 4 / 578 The train loss 7.4315123558 5 / 578 The train loss 6.93853130341 6 / 578 The train loss 6.59566688538 7 / 578 The train loss 6.66885559899 8 / 578 The train loss 6.55167818069 9 / 578 The train loss 6.36601924896 10 / 578 The train loss 6.58995685577 11 / 578 The train loss 6.44905740565 12 / 578 The train loss 6.42344820499 13 / 578 The train loss 6.18164704396 14 / 578 The train loss 6.02800767762 15 / 578 The train loss 5.82611468633 16 / 578 The train loss 5.65933337808 17 / 578 The train loss 5.52630093518 18 / 578 The train loss 5.43736101521 19 / 578 The train loss 5.34862864645 20 / 578 The train loss 5.21654934883 21 / 578 The train loss 5.08712423415 22 / 578 The train loss 4.95325220715 23 / 578 The train loss 4.84278983655 24 / 578 The train loss 4.71541872621 25 / 578 The train loss 4.64515040398 26 / 578 The train loss 4.56220798309 27 / 578 The train loss 4.45934628999 28 / 578 The train loss 4.3780022817 29 / 578 The train loss 4.29011455487 30 / 578 The train loss 4.19135899146 31 / 578 The train loss 4.11428858003 32 / 578 The train loss 4.02811782435 33 / 578 The train loss 3.94298204509 34 / 578 The train loss 3.86671134304 35 / 578 The train loss 3.79821648257 36 / 578 The train loss 3.72734497984 37 / 578 The train loss 3.66112322421 38 / 578 The train loss 3.60457846052 39 / 578 The train loss 3.54151350107 40 / 578 The train loss 3.47805323899 41 / 578 The train loss 3.42648641365 42 / 578 The train loss 3.36666142373 43 / 578 The train loss 3.31629202255 44 / 578 The train loss 3.25878701969 45 / 578 The train loss 3.21149672402 46 / 578 The train loss 3.15652722379 47 / 578 The train loss 3.1028383501 48 / 578 The train loss 3.05527851482 49 / 578 The train loss 3.00176679662 50 / 578 The train loss 2.9545246166 51 / 578 The train loss 2.91405146204 52 / 578 The train loss 2.87061388848 53 / 578 The train loss 2.83195649959 54 / 578 The train loss 2.79239652609 55 / 578 The train loss 2.75280586102 56 / 578 The train loss 2.71595588912 57 / 578 The train loss 2.67951303354 58 / 578 The train loss 2.64257676283 59 / 578 The train loss 2.6134286144 60 / 578 The train loss 2.5839203621 61 / 578 The train loss 2.55053106439 62 / 578 The train loss 2.52211697688 63 / 578 The train loss 2.49212735466 64 / 578 The train loss 2.45906356908 65 / 578 The train loss 2.42787698461 66 / 578 The train loss 2.40249097392 67 / 578 The train loss 2.37526014981 68 / 578 The train loss 2.35153020787 69 / 578 The train loss 2.32750896051 70 / 578 The train loss 2.30223340775 71 / 578 The train loss 2.27570233714 72 / 578 The train loss 2.2517717936 73 / 578 The train loss 2.22905225052 74 / 578 The train loss 2.20500161036 75 / 578 The train loss 2.18344849825 76 / 578 The train loss 2.1645661569 77 / 578 The train loss 2.14073283719 78 / 578 The train loss 2.1204605072 79 / 578 The train loss 2.10007135174 80 / 578 The train loss 2.08497132882 81 / 578 The train loss 2.06674018907 82 / 578 The train loss 2.04796554403 83 / 578 The train loss 2.02973166 84 / 578 The train loss 2.01270043424 85 / 578 The train loss 1.99327235432 86 / 578 The train loss 1.97495399519 87 / 578 The train loss 1.95561109466 88 / 578 The train loss 1.93788465451 89 / 578 The train loss 1.92304772206 90 / 578 The train loss 1.90564681855 91 / 578 The train loss 1.88928882696 92 / 578 The train loss 1.8746002608 93 / 578 The train loss 1.86003764727 94 / 578 The train loss 1.84405930372 95 / 578 The train loss 1.82831892152 96 / 578 The train loss 1.81178723493 97 / 578 The train loss 1.79617848316 98 / 578 The train loss 1.78118829885 99 / 578 The train loss 1.76636563106 100 / 578 The train loss 1.75292256892 101 / 578 The train loss 1.73976051925 102 / 578 The train loss 1.72707863824 103 / 578 The train loss 1.7130259096 104 / 578 The train loss 1.69993419573 105 / 578 The train loss 1.68875365683 106 / 578 The train loss 1.67699767396 107 / 578 The train loss 1.66666053277 108 / 578 The train loss 1.6548437948 109 / 578 The train loss 1.64245027769 110 / 578 The train loss 1.63036994528 111 / 578 The train loss 1.61977894537 112 / 578 The train loss 1.60904301011 113 / 578 The train loss 1.59962279237 114 / 578 The train loss 1.59152426793 115 / 578 The train loss 1.5839580028 116 / 578 The train loss 1.57745358708 117 / 578 The train loss 1.57179284554 118 / 578 The train loss 1.56809043329 119 / 578 The train loss 1.56520626876 120 / 578 The train loss 1.56111058046 121 / 578 The train loss 1.55796242401 122 / 578 The train loss 1.55595985348 123 / 578 The train loss 1.55583668288 124 / 578 The train loss 1.56002457055 125 / 578 The train loss 1.5642752223 126 / 578 The train loss 1.56596613167 127 / 578 The train loss 1.56889546997 128 / 578 The train loss 1.57283782074 129 / 578 The train loss 1.58000539671 130 / 578 The train loss 1.58432494906 131 / 578 The train loss 1.58627931808 132 / 578 The train loss 1.58401573562 133 / 578 The train loss 1.57937079668 134 / 578 The train loss 1.57483651122 135 / 578 The train loss 1.57375110962 136 / 578 The train loss 1.57155294892 137 / 578 The train loss 1.57040503686 138 / 578 The train loss 1.56603903269 139 / 578 The train loss 1.56193917861 140 / 578 The train loss 1.55655576331 141 / 578 The train loss 1.55237601835 142 / 578 The train loss 1.54624841751 143 / 578 The train loss 1.54211726222 144 / 578 The train loss 1.5387291991 145 / 578 The train loss 1.53596441499 146 / 578 The train loss 1.53110356894 147 / 578 The train loss 1.52548151519 148 / 578 The train loss 1.51945616184 149 / 578 The train loss 1.51302779761 150 / 578 The train loss 1.50633507888 151 / 578 The train loss 1.49845906658 152 / 578 The train loss 1.49105729339 153 / 578 The train loss 1.4840383964 154 / 578 The train loss 1.4766163944 155 / 578 The train loss 1.47058075186 156 / 578 The train loss 1.46395190156 157 / 578 The train loss 1.45759576939 158 / 578 The train loss 1.45114885195 159 / 578 The train loss 1.44477486404 160 / 578 The train loss 1.43822327573 161 / 578 The train loss 1.43168008901 162 / 578 The train loss 1.42508651537 163 / 578 The train loss 1.41835435567 164 / 578 The train loss 1.41209948172 165 / 578 The train loss 1.40550359524 166 / 578 The train loss 1.39881247892 167 / 578 The train loss 1.39258979787 168 / 578 The train loss 1.38991315485 169 / 578 The train loss 1.39154137839 170 / 578 The train loss 1.39331116431 171 / 578 The train loss 1.39679498038 172 / 578 The train loss 1.39706595935 173 / 578 The train loss 1.40148190095 174 / 578 The train loss 1.40452823865 175 / 578 The train loss 1.40930656876 176 / 578 The train loss 1.41144976663 177 / 578 The train loss 1.41172070119 178 / 578 The train loss 1.41285218784 179 / 578 The train loss 1.41511930867 180 / 578 The train loss 1.41513546142 181 / 578 The train loss 1.41442044894 182 / 578 The train loss 1.41204480123 183 / 578 The train loss 1.41111513011 184 / 578 The train loss 1.41389715121 185 / 578 The train loss 1.41852835417 186 / 578 The train loss 1.42915745929 187 / 578 The train loss 1.44071801939 188 / 578 The train loss 1.45219988106 189 / 578 The train loss 1.46205519211 190 / 578 The train loss 1.47908622183 191 / 578 The train loss 1.49172891341 192 / 578 The train loss 1.50240460628 193 / 578 The train loss 1.51075422301 194 / 578 The train loss 1.51641995237 195 / 578 The train loss 1.51893924414 196 / 578 The train loss 1.52012060157 197 / 578 The train loss 1.52642909192 198 / 578 The train loss 1.52664389123 199 / 578 The train loss 1.52766394465 200 / 578 The train loss 1.52973356098 201 / 578 The train loss 1.53595528763 202 / 578 The train loss 1.54723085124 203 / 578 The train loss 1.55658477956 204 / 578 The train loss 1.56212774533 205 / 578 The train loss 1.56583897399 206 / 578 The train loss 1.56527708543 207 / 578 The train loss 1.56415702323 208 / 578 The train loss 1.56794217601 209 / 578 The train loss 1.57598204618 210 / 578 The train loss 1.57956621335 211 / 578 The train loss 1.5834902556 212 / 578 The train loss 1.5872777051 213 / 578 The train loss 1.58747952561 214 / 578 The train loss 1.58909006514 215 / 578 The train loss 1.59037636241 216 / 578 The train loss 1.589665273 217 / 578 The train loss 1.58706966221 218 / 578 The train loss 1.58504043451 219 / 578 The train loss 1.58409477559 220 / 578 The train loss 1.5854181818 221 / 578 The train loss 1.58606492745 222 / 578 The train loss 1.58719963721 223 / 578 The train loss 1.58801084757 224 / 578 The train loss 1.59042388813 225 / 578 The train loss 1.59395369609 226 / 578 The train loss 1.59748630054 227 / 578 The train loss 1.60059439427 228 / 578 The train loss 1.60302775758 229 / 578 The train loss 1.60516188525 230 / 578 The train loss 1.60760229437 231 / 578 The train loss 1.61105502194 232 / 578 The train loss 1.61276539621 233 / 578 The train loss 1.61499060147 234 / 578 The train loss 1.6183057216 235 / 578 The train loss 1.62233623114 236 / 578 The train loss 1.62410183857 237 / 578 The train loss 1.62199404748 238 / 578 The train loss 1.61849675885 239 / 578 The train loss 1.61521708491 240 / 578 The train loss 1.61240129446 241 / 578 The train loss 1.61066764742 242 / 578 The train loss 1.60788245413 243 / 578 The train loss 1.60654847097 244 / 578 The train loss 1.60620820693 245 / 578 The train loss 1.60646115444 246 / 578 The train loss 1.60583182588 247 / 578 The train loss 1.60346735658 248 / 578 The train loss 1.60035980781 249 / 578 The train loss 1.59663574068 250 / 578 The train loss 1.59639198661 251 / 578 The train loss 1.59557349677 252 / 578 The train loss 1.59536960035 253 / 578 The train loss 1.59673715792 254 / 578 The train loss 1.60100668176 255 / 578 The train loss 1.60608840386 256 / 578 The train loss 1.61342922994 257 / 578 The train loss 1.61825293071 258 / 578 The train loss 1.62212912634 259 / 578 The train loss 1.62287039025 260 / 578 The train loss 1.62349695678 261 / 578 The train loss 1.62323668154 262 / 578 The train loss 1.62325358141 263 / 578 The train loss 1.62420270266 264 / 578 The train loss 1.62580243724 265 / 578 The train loss 1.62568876676 266 / 578 The train loss 1.62625559179 267 / 578 The train loss 1.62742886896 268 / 578 The train loss 1.63063437258 269 / 578 The train loss 1.63526451787 270 / 578 The train loss 1.6413225781 271 / 578 The train loss 1.64653332449 272 / 578 The train loss 1.65092779937 273 / 578 The train loss 1.65434647545 274 / 578 The train loss 1.65852977216 275 / 578 The train loss 1.66279484294 276 / 578 The train loss 1.66607230103 277 / 578 The train loss 1.66919050471 278 / 578 The train loss 1.67121472183 279 / 578 The train loss 1.67501051687 280 / 578 The train loss 1.67880024207 281 / 578 The train loss 1.68245847433 282 / 578 The train loss 1.68549874129 283 / 578 The train loss 1.68866348077 284 / 578 The train loss 1.69159709379 285 / 578 The train loss 1.69321569037 286 / 578 The train loss 1.69550172995 287 / 578 The train loss 1.69871312878 288 / 578 The train loss 1.7007989931 289 / 578 The train loss 1.70261672191 290 / 578 The train loss 1.70566102205 291 / 578 The train loss 1.70861118499 292 / 578 The train loss 1.71177185868 293 / 578 The train loss 1.71508688548 294 / 578 The train loss 1.71926574865 295 / 578 The train loss 1.7228454806 296 / 578 The train loss 1.72572649713 297 / 578 The train loss 1.72732898503 298 / 578 The train loss 1.72856708721 299 / 578 The train loss 1.73032234325 300 / 578 The train loss 1.73245443165 301 / 578 The train loss 1.73508488439 302 / 578 The train loss 1.7379446904 303 / 578 The train loss 1.74074678433 304 / 578 The train loss 1.74199178952 305 / 578 The train loss 1.7419733835 306 / 578 The train loss 1.74322418664 307 / 578 The train loss 1.7440018982 308 / 578 The train loss 1.74391960885 309 / 578 The train loss 1.74466441731 310 / 578 The train loss 1.74272715449 311 / 578 The train loss 1.73992620308 312 / 578 The train loss 1.73698647129 313 / 578 The train loss 1.73394999184 314 / 578 The train loss 1.73160531358 315 / 578 The train loss 1.72904378989 316 / 578 The train loss 1.72625914291 317 / 578 The train loss 1.72384193511 318 / 578 The train loss 1.72126015458 319 / 578 The train loss 1.71866844515 320 / 578 The train loss 1.71554709505 321 / 578 The train loss 1.71189262098 322 / 578 The train loss 1.70816414104 323 / 578 The train loss 1.70459793983 324 / 578 The train loss 1.70035077061 325 / 578 The train loss 1.69669572665 326 / 578 The train loss 1.69305280271 327 / 578 The train loss 1.68904889487 328 / 578 The train loss 1.68524367526 329 / 578 The train loss 1.6818154976 330 / 578 The train loss 1.6782347419 331 / 578 The train loss 1.67451861024 332 / 578 The train loss 1.67120331741 333 / 578 The train loss 1.66767937786 334 / 578 The train loss 1.66534419545 335 / 578 The train loss 1.66373571851 336 / 578 The train loss 1.66304894501 337 / 578 The train loss 1.66652179472 338 / 578 The train loss 1.66991892016 339 / 578 The train loss 1.67356912858 340 / 578 The train loss 1.67401372685 341 / 578 The train loss 1.67259796204 342 / 578 The train loss 1.67021759979 343 / 578 The train loss 1.66769005345 344 / 578 The train loss 1.66454617766 345 / 578 The train loss 1.66243985459 346 / 578 The train loss 1.66073616805 347 / 578 The train loss 1.65925126564 348 / 578 The train loss 1.65784200409 349 / 578 The train loss 1.65683290712 350 / 578 The train loss 1.65574410609 351 / 578 The train loss 1.6553706053 352 / 578 The train loss 1.65435549888 353 / 578 The train loss 1.65272055672 354 / 578 The train loss 1.65137479029 355 / 578 The train loss 1.65078886865 356 / 578 The train loss 1.64986623773 357 / 578 The train loss 1.64920790155 358 / 578 The train loss 1.64764642482 359 / 578 The train loss 1.64592838354 360 / 578 The train loss 1.64428686963 361 / 578 The train loss 1.64310649128 362 / 578 The train loss 1.64135440937 363 / 578 The train loss 1.63912331335 364 / 578 The train loss 1.63679510289 365 / 578 The train loss 1.63479098229 366 / 578 The train loss 1.63115485004 367 / 578 The train loss 1.62716210039 368 / 578 The train loss 1.62320837221 369 / 578 The train loss 1.61916137473 370 / 578 The train loss 1.61530822689 371 / 578 The train loss 1.61142583802 372 / 578 The train loss 1.60762735821 373 / 578 The train loss 1.60381411411 374 / 578 The train loss 1.60028071518 375 / 578 The train loss 1.59658535326 376 / 578 The train loss 1.59283816656 377 / 578 The train loss 1.58911585468 378 / 578 The train loss 1.58555990852 379 / 578 The train loss 1.58200384275 380 / 578 The train loss 1.57838699061 381 / 578 The train loss 1.57530177817 382 / 578 The train loss 1.57233494633 383 / 578 The train loss 1.56953910207 384 / 578 The train loss 1.56644503349 385 / 578 The train loss 1.56331716524 386 / 578 The train loss 1.55974480637 387 / 578 The train loss 1.55652300498 388 / 578 The train loss 1.55339701961 389 / 578 The train loss 1.55027646466 390 / 578 The train loss 1.54689440139 391 / 578 The train loss 1.54359028002 392 / 578 The train loss 1.5400317011 393 / 578 The train loss 1.53672817634 394 / 578 The train loss 1.53338239913 395 / 578 The train loss 1.52982067436 396 / 578 The train loss 1.52691767244 397 / 578 The train loss 1.52434869009 398 / 578 The train loss 1.52187004633 399 / 578 The train loss 1.51915259521 400 / 578 The train loss 1.51636760511 401 / 578 The train loss 1.51386732406 402 / 578 The train loss 1.51119429511 403 / 578 The train loss 1.50924373708 404 / 578 The train loss 1.50709120708 405 / 578 The train loss 1.50504031181 406 / 578 The train loss 1.50291537211 407 / 578 The train loss 1.50119139477 408 / 578 The train loss 1.49933238459 409 / 578 The train loss 1.49971129128 410 / 578 The train loss 1.50214732028 411 / 578 The train loss 1.50365772836 412 / 578 The train loss 1.50415666083 413 / 578 The train loss 1.50398789783 414 / 578 The train loss 1.50482310952 415 / 578 The train loss 1.50508089022 416 / 578 The train loss 1.50404201319 417 / 578 The train loss 1.50294981057 418 / 578 The train loss 1.50181968831 419 / 578 The train loss 1.50070634363 420 / 578 The train loss 1.49987229095 421 / 578 The train loss 1.49898811148 422 / 578 The train loss 1.49761489768 423 / 578 The train loss 1.49680900588 424 / 578 The train loss 1.49648200496 425 / 578 The train loss 1.49503165231 426 / 578 The train loss 1.49346619257 427 / 578 The train loss 1.49173541147 428 / 578 The train loss 1.48970105186 429 / 578 The train loss 1.48763140432 430 / 578 The train loss 1.48498695742 431 / 578 The train loss 1.48217968383 432 / 578 The train loss 1.47948264883 433 / 578 The train loss 1.47647907559 434 / 578 The train loss 1.47338039943 435 / 578 The train loss 1.47034600319 436 / 578 The train loss 1.4673167261 437 / 578 The train loss 1.46454636255 438 / 578 The train loss 1.4622768162 439 / 578 The train loss 1.46015650058 440 / 578 The train loss 1.45811235969 441 / 578 The train loss 1.45582415522 442 / 578 The train loss 1.45349500821 443 / 578 The train loss 1.45136192269 444 / 578 The train loss 1.45011834757 445 / 578 The train loss 1.44913978654 446 / 578 The train loss 1.44849116596 447 / 578 The train loss 1.44785448392 448 / 578 The train loss 1.44682076158 449 / 578 The train loss 1.44592591951 450 / 578 The train loss 1.44456569377 451 / 578 The train loss 1.44361492849 452 / 578 The train loss 1.44253811 453 / 578 The train loss 1.44118113651 454 / 578 The train loss 1.44008326002 455 / 578 The train loss 1.43850890591 456 / 578 The train loss 1.43684305012 457 / 578 The train loss 1.43505518791 458 / 578 The train loss 1.43304480642 459 / 578 The train loss 1.4313137453 460 / 578 The train loss 1.42975519828 461 / 578 The train loss 1.42811736688 462 / 578 The train loss 1.4261994328 463 / 578 The train loss 1.42464192379 464 / 578 The train loss 1.42221841052 465 / 578 The train loss 1.41981585817 466 / 578 The train loss 1.41745148315 467 / 578 The train loss 1.41498010767 468 / 578 The train loss 1.4125803156 469 / 578 The train loss 1.41032391523 470 / 578 The train loss 1.40812069916 471 / 578 The train loss 1.40606255677 472 / 578 The train loss 1.40386548654 473 / 578 The train loss 1.4017878039 474 / 578 The train loss 1.39952696064 475 / 578 The train loss 1.39727523556 476 / 578 The train loss 1.3951733457 477 / 578 The train loss 1.39293857331 478 / 578 The train loss 1.39052790023 479 / 578 The train loss 1.38800881352 480 / 578 The train loss 1.38555215579 481 / 578 The train loss 1.38312507314 482 / 578 The train loss 1.38056506795 483 / 578 The train loss 1.3780954926 484 / 578 The train loss 1.37558570292 485 / 578 The train loss 1.37320695873 486 / 578 The train loss 1.37089319021 487 / 578 The train loss 1.36844719016 488 / 578 The train loss 1.36609858784 489 / 578 The train loss 1.36368053076 490 / 578 The train loss 1.36132111726 491 / 578 The train loss 1.35888813527 492 / 578 The train loss 1.35648856557 493 / 578 The train loss 1.35418054981 494 / 578 The train loss 1.35174288368 495 / 578 The train loss 1.34961194715 496 / 578 The train loss 1.34766892314 497 / 578 The train loss 1.34580954358 498 / 578 The train loss 1.34402044871 499 / 578 The train loss 1.34209804359 500 / 578 The train loss 1.34002829021 501 / 578 The train loss 1.33780091867 502 / 578 The train loss 1.33559197036 503 / 578 The train loss 1.33363583906 504 / 578 The train loss 1.33163740181 505 / 578 The train loss 1.3296899165 506 / 578 The train loss 1.32758306364 507 / 578 The train loss 1.32545600717 508 / 578 The train loss 1.32313637875 509 / 578 The train loss 1.32078089534 510 / 578 The train loss 1.31846296001 511 / 578 The train loss 1.31624912694 512 / 578 The train loss 1.31399258049 513 / 578 The train loss 1.31168094916 514 / 578 The train loss 1.30937190115 515 / 578 The train loss 1.30711896901 516 / 578 The train loss 1.3049591818 517 / 578 The train loss 1.30284423762 518 / 578 The train loss 1.30069711777 519 / 578 The train loss 1.29861703086 520 / 578 The train loss 1.29649998874 521 / 578 The train loss 1.29445909303 522 / 578 The train loss 1.29219766419 523 / 578 The train loss 1.29019641331 524 / 578 The train loss 1.28847568649 525 / 578 The train loss 1.28651582997 526 / 578 The train loss 1.28475477522 527 / 578 The train loss 1.28292782535 528 / 578 The train loss 1.28130720337 529 / 578 The train loss 1.27980938978 530 / 578 The train loss 1.27820918342 531 / 578 The train loss 1.27647934482 532 / 578 The train loss 1.27480241652 533 / 578 The train loss 1.27309768566 534 / 578 The train loss 1.2712262816 535 / 578 The train loss 1.26932260462 536 / 578 The train loss 1.26739740184 537 / 578 The train loss 1.26548258634 538 / 578 The train loss 1.26363289859 539 / 578 The train loss 1.26186459011 540 / 578 The train loss 1.26011506588 541 / 578 The train loss 1.25840308721 542 / 578 The train loss 1.25691020702 543 / 578 The train loss 1.25518298894 544 / 578 The train loss 1.25366067299 545 / 578 The train loss 1.25242952277 546 / 578 The train loss 1.25226567499 547 / 578 The train loss 1.25264215931 548 / 578 The train loss 1.25445160535 549 / 578 The train loss 1.25497647695 550 / 578 The train loss 1.25474603679 551 / 578 The train loss 1.25435364056 552 / 578 The train loss 1.25391982975 553 / 578 The train loss 1.25355170307 554 / 578 The train loss 1.25322358944 555 / 578 The train loss 1.25241886972 556 / 578 The train loss 1.25157541169 557 / 578 The train loss 1.25092592171 558 / 578 The train loss 1.25022478561 559 / 578 The train loss 1.24963046084 560 / 578 The train loss 1.24915469644 561 / 578 The train loss 1.24862399622 562 / 578 The train loss 1.24810766647 563 / 578 The train loss 1.24737608225 564 / 578 The train loss 1.24585133288 565 / 578 The train loss 1.24423871081 566 / 578 The train loss 1.24250777844 567 / 578 The train loss 1.24086058727 568 / 578 The train loss 1.23930011956 569 / 578 The train loss 1.23762003047 570 / 578 The train loss 1.23588562099 571 / 578 The train loss 1.23416721491 572 / 578 The train loss 1.23271814425 573 / 578 The train loss 1.23215758019 574 / 578 The train loss 1.23145601246 575 / 578 The train loss 1.23027329744 576 / 578 The train loss 1.22933373692 577 / 578 The train loss 1.22837022591 578 / 578 The train loss 1.22749360136 Starting epoch 1 Validation: 1 / 30 The valid loss 0.461220532656 2 / 30 The valid loss 0.435441240668 3 / 30 The valid loss 0.413019051154 4 / 30 The valid loss 0.388976730406 5 / 30 The valid loss 0.368342584372 6 / 30 The valid loss 0.407765770952 7 / 30 The valid loss 0.521570056677 8 / 30 The valid loss 0.66619252041 9 / 30 The valid loss 0.813994119565 10 / 30 The valid loss 0.894038441777 11 / 30 The valid loss 0.911667046222 12 / 30 The valid loss 0.922555280228 13 / 30 The valid loss 0.93134905971 14 / 30 The valid loss 0.936072607126 15 / 30 The valid loss 0.919061885277 16 / 30 The valid loss 0.911798747256 17 / 30 The valid loss 0.924880331053 18 / 30 The valid loss 0.934616315696 19 / 30 The valid loss 0.934962769872 20 / 30 The valid loss 0.928810422122 21 / 30 The valid loss 0.914856412581 22 / 30 The valid loss 0.901876870881 23 / 30 The valid loss 0.890152232802 24 / 30 The valid loss 0.881474620352 25 / 30 The valid loss 0.873161503077 26 / 30 The valid loss 0.869586436794 27 / 30 The valid loss 0.858789041086 28 / 30 The valid loss 0.839347845742 29 / 30 The valid loss 0.817362181072 30 / 30 The valid loss 0.79903665185 Validation MSE(val_loss): 16.9458305262 Test MSE(test_loss): 5.62044679525 Training 1 / 578 The train loss 0.656893014908 2 / 578 The train loss 0.51439127326 3 / 578 The train loss 0.450964212418 4 / 578 The train loss 0.425492420793 5 / 578 The train loss 0.399875622988 6 / 578 The train loss 0.363117776811 7 / 578 The train loss 0.335625186563 8 / 578 The train loss 0.315836992115 9 / 578 The train loss 0.292885882159 10 / 578 The train loss 0.277139987797 11 / 578 The train loss 0.264178500257 12 / 578 The train loss 0.255386890844 13 / 578 The train loss 0.251184011308 14 / 578 The train loss 0.258106558983 15 / 578 The train loss 0.260125276943 16 / 578 The train loss 0.259601559024 17 / 578 The train loss 0.257349094486 18 / 578 The train loss 0.251266757233 19 / 578 The train loss 0.244739772066 20 / 578 The train loss 0.238786169514 21 / 578 The train loss 0.232277642758 22 / 578 The train loss 0.227042474869 23 / 578 The train loss 0.221370988566 24 / 578 The train loss 0.216529093062 25 / 578 The train loss 0.213741371036 26 / 578 The train loss 0.209614010384 27 / 578 The train loss 0.206228588742 28 / 578 The train loss 0.204224795369 29 / 578 The train loss 0.202027257917 30 / 578 The train loss 0.201041421046 31 / 578 The train loss 0.199986364332 32 / 578 The train loss 0.196609286126 33 / 578 The train loss 0.19690700643 34 / 578 The train loss 0.197788525154 35 / 578 The train loss 0.201017017875 36 / 578 The train loss 0.206119137506 37 / 578 The train loss 0.211359546797 38 / 578 The train loss 0.215108318548 39 / 578 The train loss 0.216912112939 40 / 578 The train loss 0.217340109125 41 / 578 The train loss 0.217207977321 42 / 578 The train loss 0.216916013332 43 / 578 The train loss 0.218226962311 44 / 578 The train loss 0.219618033279 45 / 578 The train loss 0.219488375054 46 / 578 The train loss 0.218842970288 47 / 578 The train loss 0.217847958524 48 / 578 The train loss 0.21547418957 49 / 578 The train loss 0.214290551689 50 / 578 The train loss 0.215696783364 51 / 578 The train loss 0.217336607622 52 / 578 The train loss 0.218742664903 53 / 578 The train loss 0.220638734833 54 / 578 The train loss 0.220838974747 55 / 578 The train loss 0.219859360294 56 / 578 The train loss 0.219091082258 57 / 578 The train loss 0.22038644134 58 / 578 The train loss 0.221880807959 59 / 578 The train loss 0.22574362967 60 / 578 The train loss 0.231338033577 61 / 578 The train loss 0.240988778775 62 / 578 The train loss 0.250158797829 63 / 578 The train loss 0.259307963981 64 / 578 The train loss 0.265895504039 65 / 578 The train loss 0.270350252207 66 / 578 The train loss 0.272202163935 67 / 578 The train loss 0.272752081725 68 / 578 The train loss 0.272967769819 69 / 578 The train loss 0.274060059285 70 / 578 The train loss 0.277240884304 71 / 578 The train loss 0.280925093402 72 / 578 The train loss 0.28644066304 73 / 578 The train loss 0.289900208173 74 / 578 The train loss 0.293082596482 75 / 578 The train loss 0.295621138016 76 / 578 The train loss 0.296998007125 77 / 578 The train loss 0.298964181504 78 / 578 The train loss 0.301299771437 79 / 578 The train loss 0.303168685753 80 / 578 The train loss 0.304361683503 81 / 578 The train loss 0.304352026663 82 / 578 The train loss 0.303713121065 83 / 578 The train loss 0.303161279983 84 / 578 The train loss 0.303085266125 85 / 578 The train loss 0.301937159195 86 / 578 The train loss 0.300288227235 87 / 578 The train loss 0.298835376727 88 / 578 The train loss 0.29782445932 89 / 578 The train loss 0.298577534684 90 / 578 The train loss 0.30013079279 91 / 578 The train loss 0.30141972644 92 / 578 The train loss 0.3014316967 93 / 578 The train loss 0.300915666806 94 / 578 The train loss 0.299373526205 95 / 578 The train loss 0.297749148231 96 / 578 The train loss 0.297088483814 97 / 578 The train loss 0.296877290017 98 / 578 The train loss 0.298118478942 99 / 578 The train loss 0.300299087709 100 / 578 The train loss 0.303118702322 101 / 578 The train loss 0.304696338159 102 / 578 The train loss 0.303872570542 103 / 578 The train loss 0.303453248537 104 / 578 The train loss 0.303928750782 105 / 578 The train loss 0.304136334005 106 / 578 The train loss 0.304696375891 107 / 578 The train loss 0.305056565137 108 / 578 The train loss 0.30688495865 109 / 578 The train loss 0.309691306511 110 / 578 The train loss 0.309523866258 111 / 578 The train loss 0.311020504113 112 / 578 The train loss 0.311792281988 113 / 578 The train loss 0.312045487814 114 / 578 The train loss 0.313517398087 115 / 578 The train loss 0.314814945796 116 / 578 The train loss 0.317132362389 117 / 578 The train loss 0.320111803902 118 / 578 The train loss 0.323943573413 119 / 578 The train loss 0.329404630456 120 / 578 The train loss 0.333288448925 121 / 578 The train loss 0.338159763493 122 / 578 The train loss 0.343286926018 123 / 578 The train loss 0.351044017488 124 / 578 The train loss 0.361292103005 125 / 578 The train loss 0.371584941268 126 / 578 The train loss 0.38116372424 127 / 578 The train loss 0.392086866686 128 / 578 The train loss 0.402758778422 129 / 578 The train loss 0.414625478692 130 / 578 The train loss 0.428024305174 131 / 578 The train loss 0.43787434299 132 / 578 The train loss 0.442678477164 133 / 578 The train loss 0.444932037614 134 / 578 The train loss 0.446516144076 135 / 578 The train loss 0.450106933271 136 / 578 The train loss 0.455187159317 137 / 578 The train loss 0.460371424573 138 / 578 The train loss 0.463160760485 139 / 578 The train loss 0.4658042007 140 / 578 The train loss 0.46705431076 141 / 578 The train loss 0.467391615113 142 / 578 The train loss 0.467903237842 143 / 578 The train loss 0.469614225563 144 / 578 The train loss 0.472060795356 145 / 578 The train loss 0.47324827447 146 / 578 The train loss 0.474017959241 147 / 578 The train loss 0.474011410357 148 / 578 The train loss 0.473725590654 149 / 578 The train loss 0.473600581088 150 / 578 The train loss 0.47412418435 151 / 578 The train loss 0.472671338955 152 / 578 The train loss 0.47134429736 153 / 578 The train loss 0.470456857794 154 / 578 The train loss 0.46946498471 155 / 578 The train loss 0.469528264865 156 / 578 The train loss 0.469928327757 157 / 578 The train loss 0.470833854406 158 / 578 The train loss 0.472257232081 159 / 578 The train loss 0.473017053128 160 / 578 The train loss 0.473097492103 161 / 578 The train loss 0.472300076429 162 / 578 The train loss 0.47242046957 163 / 578 The train loss 0.472602029245 164 / 578 The train loss 0.472480124544 165 / 578 The train loss 0.471730420355 166 / 578 The train loss 0.47144219773 167 / 578 The train loss 0.471188934888 168 / 578 The train loss 0.473648665828 169 / 578 The train loss 0.478795296254 170 / 578 The train loss 0.486863739964 171 / 578 The train loss 0.49397994543 172 / 578 The train loss 0.499667085794 173 / 578 The train loss 0.506144971431 174 / 578 The train loss 0.51438918804 175 / 578 The train loss 0.520788541436 176 / 578 The train loss 0.527301016094 177 / 578 The train loss 0.531811522815 178 / 578 The train loss 0.535075665106 179 / 578 The train loss 0.538823095031 180 / 578 The train loss 0.542200564593 181 / 578 The train loss 0.544558169351 182 / 578 The train loss 0.545623917419 183 / 578 The train loss 0.546993440536 184 / 578 The train loss 0.551764495914 185 / 578 The train loss 0.557521292889 186 / 578 The train loss 0.570879860511 187 / 578 The train loss 0.588565629992 188 / 578 The train loss 0.60379695504 189 / 578 The train loss 0.616499372181 190 / 578 The train loss 0.632003413925 191 / 578 The train loss 0.644923282465 192 / 578 The train loss 0.656842821511 193 / 578 The train loss 0.667112529509 194 / 578 The train loss 0.67460159535 195 / 578 The train loss 0.678933520118 196 / 578 The train loss 0.683704879135 197 / 578 The train loss 0.691371891629 198 / 578 The train loss 0.695130027229 199 / 578 The train loss 0.699180781467 200 / 578 The train loss 0.704269750044 201 / 578 The train loss 0.713848593149 202 / 578 The train loss 0.728205827129 203 / 578 The train loss 0.740272743875 204 / 578 The train loss 0.749833096403 205 / 578 The train loss 0.756452010245 206 / 578 The train loss 0.75975119771 207 / 578 The train loss 0.762821278062 208 / 578 The train loss 0.771033676699 209 / 578 The train loss 0.782010526677 210 / 578 The train loss 0.78999755489 211 / 578 The train loss 0.796734639183 212 / 578 The train loss 0.803386886095 213 / 578 The train loss 0.809608458283 214 / 578 The train loss 0.814040996691 215 / 578 The train loss 0.817935486658 216 / 578 The train loss 0.821032472714 217 / 578 The train loss 0.822292116312 218 / 578 The train loss 0.824016759087 219 / 578 The train loss 0.826812064743 220 / 578 The train loss 0.829710317674 221 / 578 The train loss 0.833871347804 222 / 578 The train loss 0.837919125082 223 / 578 The train loss 0.842254639674 224 / 578 The train loss 0.846780111681 225 / 578 The train loss 0.854029669166 226 / 578 The train loss 0.861799362421 227 / 578 The train loss 0.869124710757 228 / 578 The train loss 0.874426386793 229 / 578 The train loss 0.880692478806 230 / 578 The train loss 0.886911554894 231 / 578 The train loss 0.893217324114 232 / 578 The train loss 0.898513766134 233 / 578 The train loss 0.904062571891 234 / 578 The train loss 0.910720734451 235 / 578 The train loss 0.91922141434 236 / 578 The train loss 0.924496851672 237 / 578 The train loss 0.925650761537 238 / 578 The train loss 0.925076751586 239 / 578 The train loss 0.924946717076 240 / 578 The train loss 0.925549200612 241 / 578 The train loss 0.927205178191 242 / 578 The train loss 0.929632509481 243 / 578 The train loss 0.932009726334 244 / 578 The train loss 0.93537494583 245 / 578 The train loss 0.940716593545 246 / 578 The train loss 0.945506816896 247 / 578 The train loss 0.947076435453 248 / 578 The train loss 0.948605849317 249 / 578 The train loss 0.950703143535 250 / 578 The train loss 0.954853717744 251 / 578 The train loss 0.959139605383 252 / 578 The train loss 0.963885344151 253 / 578 The train loss 0.970825473252 254 / 578 The train loss 0.980385698613 255 / 578 The train loss 0.990481887086 256 / 578 The train loss 1.00271268183 257 / 578 The train loss 1.01382757341 258 / 578 The train loss 1.02330581064 259 / 578 The train loss 1.0288149417 260 / 578 The train loss 1.03408712877 261 / 578 The train loss 1.03940809178 262 / 578 The train loss 1.04420892058 263 / 578 The train loss 1.04900441327 264 / 578 The train loss 1.05547956168 265 / 578 The train loss 1.06088992239 266 / 578 The train loss 1.06731364865 267 / 578 The train loss 1.07405857999 268 / 578 The train loss 1.08228694764 269 / 578 The train loss 1.09176159077 270 / 578 The train loss 1.10355501346 271 / 578 The train loss 1.11349708474 272 / 578 The train loss 1.12236233576 273 / 578 The train loss 1.13164720862 274 / 578 The train loss 1.14075772861 275 / 578 The train loss 1.14955613911 276 / 578 The train loss 1.15788460636 277 / 578 The train loss 1.16475572563 278 / 578 The train loss 1.1724445159 279 / 578 The train loss 1.18026555904 280 / 578 The train loss 1.18956363622 281 / 578 The train loss 1.19952603331 282 / 578 The train loss 1.20770939524 283 / 578 The train loss 1.21626285103 284 / 578 The train loss 1.22478228382 285 / 578 The train loss 1.23277425426 286 / 578 The train loss 1.2400544081 287 / 578 The train loss 1.24737174414 288 / 578 The train loss 1.25457898771 289 / 578 The train loss 1.26163235993 290 / 578 The train loss 1.26867332125 291 / 578 The train loss 1.27635123225 292 / 578 The train loss 1.28502020684 293 / 578 The train loss 1.2928420911 294 / 578 The train loss 1.29987602407 295 / 578 The train loss 1.3072320679 296 / 578 The train loss 1.313424882 297 / 578 The train loss 1.31888879665 298 / 578 The train loss 1.32399127142 299 / 578 The train loss 1.32966924745 300 / 578 The train loss 1.33653313314 301 / 578 The train loss 1.34251298614 302 / 578 The train loss 1.34883016067 303 / 578 The train loss 1.35458592215 304 / 578 The train loss 1.35986470701 305 / 578 The train loss 1.36442580765 306 / 578 The train loss 1.36847371053 307 / 578 The train loss 1.37293585577 308 / 578 The train loss 1.37721237729 309 / 578 The train loss 1.38024585142 310 / 578 The train loss 1.38091623701 311 / 578 The train loss 1.37995887411 312 / 578 The train loss 1.37798455658 313 / 578 The train loss 1.375754948 314 / 578 The train loss 1.37392989612 315 / 578 The train loss 1.37254826072 316 / 578 The train loss 1.37158854584 317 / 578 The train loss 1.37096722866 318 / 578 The train loss 1.37009012994 319 / 578 The train loss 1.36895679152 320 / 578 The train loss 1.36688739671 321 / 578 The train loss 1.36376420309 322 / 578 The train loss 1.36048505779 323 / 578 The train loss 1.35702666031 324 / 578 The train loss 1.35349482583 325 / 578 The train loss 1.3499563816 326 / 578 The train loss 1.34622567046 327 / 578 The train loss 1.34252748007 328 / 578 The train loss 1.33879218344 329 / 578 The train loss 1.33537965137 330 / 578 The train loss 1.33186484108 331 / 578 The train loss 1.3283430683 332 / 578 The train loss 1.32485364891 333 / 578 The train loss 1.32164049357 334 / 578 The train loss 1.31927538768 335 / 578 The train loss 1.31800700117 336 / 578 The train loss 1.31752034483 337 / 578 The train loss 1.31785035551 338 / 578 The train loss 1.31894520465 339 / 578 The train loss 1.31963977484 340 / 578 The train loss 1.31863917842 341 / 578 The train loss 1.31628848234 342 / 578 The train loss 1.313835583 343 / 578 The train loss 1.31140301068 344 / 578 The train loss 1.3091624025 345 / 578 The train loss 1.30718124943 346 / 578 The train loss 1.30512872556 347 / 578 The train loss 1.30358159038 348 / 578 The train loss 1.30228859351 349 / 578 The train loss 1.30151631286 350 / 578 The train loss 1.30095872498 351 / 578 The train loss 1.30016417884 352 / 578 The train loss 1.29950726839 353 / 578 The train loss 1.29861430839 354 / 578 The train loss 1.29801577862 355 / 578 The train loss 1.29733872172 356 / 578 The train loss 1.29696224138 357 / 578 The train loss 1.2967451523 358 / 578 The train loss 1.29639168419 359 / 578 The train loss 1.29541828218 360 / 578 The train loss 1.29454967854 361 / 578 The train loss 1.29372138274 362 / 578 The train loss 1.29250973161 363 / 578 The train loss 1.29102756918 364 / 578 The train loss 1.28935748857 365 / 578 The train loss 1.28774967579 366 / 578 The train loss 1.28458150881 367 / 578 The train loss 1.28129503267 368 / 578 The train loss 1.27805851031 369 / 578 The train loss 1.27476920292 370 / 578 The train loss 1.2715237578 371 / 578 The train loss 1.26840277346 372 / 578 The train loss 1.26530045822 373 / 578 The train loss 1.2622729119 374 / 578 The train loss 1.25920696186 375 / 578 The train loss 1.25625469522 376 / 578 The train loss 1.25329092704 377 / 578 The train loss 1.25033488423 378 / 578 The train loss 1.24736317302 379 / 578 The train loss 1.24446725004 380 / 578 The train loss 1.24167561562 381 / 578 The train loss 1.23915055652 382 / 578 The train loss 1.23687944729 383 / 578 The train loss 1.23455453254 384 / 578 The train loss 1.23215456881 385 / 578 The train loss 1.22966956458 386 / 578 The train loss 1.22694032678 387 / 578 The train loss 1.22419191235 388 / 578 The train loss 1.22147807694 389 / 578 The train loss 1.21892908376 390 / 578 The train loss 1.21615725985 391 / 578 The train loss 1.2133354242 392 / 578 The train loss 1.21055522869 393 / 578 The train loss 1.20780301598 394 / 578 The train loss 1.2049866223 395 / 578 The train loss 1.20217246364 396 / 578 The train loss 1.19985928902 397 / 578 The train loss 1.19785334823 398 / 578 The train loss 1.19624417912 399 / 578 The train loss 1.19449371275 400 / 578 The train loss 1.19274387863 401 / 578 The train loss 1.19093902223 402 / 578 The train loss 1.18924003762 403 / 578 The train loss 1.18786574412 404 / 578 The train loss 1.18647546782 405 / 578 The train loss 1.1853338874 406 / 578 The train loss 1.18411819567 407 / 578 The train loss 1.18304497674 408 / 578 The train loss 1.1819630224 409 / 578 The train loss 1.18349276017 410 / 578 The train loss 1.18691443986 411 / 578 The train loss 1.18906784982 412 / 578 The train loss 1.1900777666 413 / 578 The train loss 1.19081665323 414 / 578 The train loss 1.19210467761 415 / 578 The train loss 1.19294096127 416 / 578 The train loss 1.19271716071 417 / 578 The train loss 1.19205286216 418 / 578 The train loss 1.19168316102 419 / 578 The train loss 1.19154257119 420 / 578 The train loss 1.19157863667 421 / 578 The train loss 1.19173449048 422 / 578 The train loss 1.19164807575 423 / 578 The train loss 1.19183845001 424 / 578 The train loss 1.19170311863 425 / 578 The train loss 1.19129520343 426 / 578 The train loss 1.19034975652 427 / 578 The train loss 1.18910317646 428 / 578 The train loss 1.18800997814 429 / 578 The train loss 1.18641946724 430 / 578 The train loss 1.18453396909 431 / 578 The train loss 1.18242623832 432 / 578 The train loss 1.18025222679 433 / 578 The train loss 1.17800650713 434 / 578 The train loss 1.17558433734 435 / 578 The train loss 1.1731749239 436 / 578 The train loss 1.17092702042 437 / 578 The train loss 1.16888041167 438 / 578 The train loss 1.16742349382 439 / 578 The train loss 1.16630469276 440 / 578 The train loss 1.16518949436 441 / 578 The train loss 1.16389926656 442 / 578 The train loss 1.16246276814 443 / 578 The train loss 1.16129499368 444 / 578 The train loss 1.16085128664 445 / 578 The train loss 1.16100060782 446 / 578 The train loss 1.16161284161 447 / 578 The train loss 1.16226709719 448 / 578 The train loss 1.16279007099 449 / 578 The train loss 1.16279425869 450 / 578 The train loss 1.1625788832 451 / 578 The train loss 1.16261007229 452 / 578 The train loss 1.16243694133 453 / 578 The train loss 1.16200727807 454 / 578 The train loss 1.16136773282 455 / 578 The train loss 1.16105873247 456 / 578 The train loss 1.16058595194 457 / 578 The train loss 1.1598514924 458 / 578 The train loss 1.15901497478 459 / 578 The train loss 1.15798786526 460 / 578 The train loss 1.15752958788 461 / 578 The train loss 1.15667505588 462 / 578 The train loss 1.15581868934 463 / 578 The train loss 1.15455914628 464 / 578 The train loss 1.15266270065 465 / 578 The train loss 1.15068500971 466 / 578 The train loss 1.14846447809 467 / 578 The train loss 1.14635546501 468 / 578 The train loss 1.14440328184 469 / 578 The train loss 1.14265673217 470 / 578 The train loss 1.1407789351 471 / 578 The train loss 1.13901095192 472 / 578 The train loss 1.13704468046 473 / 578 The train loss 1.13521015362 474 / 578 The train loss 1.13331929351 475 / 578 The train loss 1.13132935334 476 / 578 The train loss 1.12954833689 477 / 578 The train loss 1.12765552111 478 / 578 The train loss 1.1256697674 479 / 578 The train loss 1.12349100926 480 / 578 The train loss 1.12135039736 481 / 578 The train loss 1.11926238883 482 / 578 The train loss 1.11715824455 483 / 578 The train loss 1.11508427594 484 / 578 The train loss 1.11307002142 485 / 578 The train loss 1.11113490536 486 / 578 The train loss 1.10917745138 487 / 578 The train loss 1.10719293561 488 / 578 The train loss 1.1053315122 489 / 578 The train loss 1.10336438211 490 / 578 The train loss 1.10136573765 491 / 578 The train loss 1.0995356229 492 / 578 The train loss 1.09761565642 493 / 578 The train loss 1.0957015157 494 / 578 The train loss 1.09373321555 495 / 578 The train loss 1.09201102604 496 / 578 The train loss 1.09045896948 497 / 578 The train loss 1.08905314066 498 / 578 The train loss 1.08758818647 499 / 578 The train loss 1.08603458221 500 / 578 The train loss 1.08427067368 501 / 578 The train loss 1.08246826053 502 / 578 The train loss 1.08082322917 503 / 578 The train loss 1.07920881826 504 / 578 The train loss 1.07763928014 505 / 578 The train loss 1.07596923849 506 / 578 The train loss 1.07434106502 507 / 578 The train loss 1.07290328406 508 / 578 The train loss 1.07117021534 509 / 578 The train loss 1.06936819414 510 / 578 The train loss 1.06752316737 511 / 578 The train loss 1.06565786828 512 / 578 The train loss 1.06386256317 513 / 578 The train loss 1.06196459117 514 / 578 The train loss 1.06016278157 515 / 578 The train loss 1.05840609704 516 / 578 The train loss 1.05667989337 517 / 578 The train loss 1.05508447875 518 / 578 The train loss 1.0534417135 519 / 578 The train loss 1.05169803289 520 / 578 The train loss 1.05009503021 521 / 578 The train loss 1.04845964645 522 / 578 The train loss 1.0466845541 523 / 578 The train loss 1.04499381252 524 / 578 The train loss 1.04343014474 525 / 578 The train loss 1.04201297472 526 / 578 The train loss 1.04059812043 527 / 578 The train loss 1.03926926361 528 / 578 The train loss 1.03793602912 529 / 578 The train loss 1.03675731257 530 / 578 The train loss 1.03541422685 531 / 578 The train loss 1.03405722708 532 / 578 The train loss 1.03280507388 533 / 578 The train loss 1.03147959532 534 / 578 The train loss 1.02999813206 535 / 578 The train loss 1.02842283364 536 / 578 The train loss 1.0269291249 537 / 578 The train loss 1.02540784693 538 / 578 The train loss 1.02409214381 539 / 578 The train loss 1.02265296836 540 / 578 The train loss 1.0212907969 541 / 578 The train loss 1.02000726488 542 / 578 The train loss 1.01873620879 543 / 578 The train loss 1.01740749766 544 / 578 The train loss 1.01618254137 545 / 578 The train loss 1.01522224782 546 / 578 The train loss 1.01544123955 547 / 578 The train loss 1.01632642311 548 / 578 The train loss 1.01863299018 549 / 578 The train loss 1.01983500267 550 / 578 The train loss 1.02016944735 551 / 578 The train loss 1.02045104574 552 / 578 The train loss 1.02047475094 553 / 578 The train loss 1.02040753605 554 / 578 The train loss 1.02032388151 555 / 578 The train loss 1.01992375476 556 / 578 The train loss 1.01950145895 557 / 578 The train loss 1.01921487145 558 / 578 The train loss 1.01880454474 559 / 578 The train loss 1.01852468292 560 / 578 The train loss 1.01840490828 561 / 578 The train loss 1.01816697226 562 / 578 The train loss 1.01802615839 563 / 578 The train loss 1.01757349057 564 / 578 The train loss 1.01629269817 565 / 578 The train loss 1.01500532127 566 / 578 The train loss 1.01364267935 567 / 578 The train loss 1.01226791373 568 / 578 The train loss 1.01095019723 569 / 578 The train loss 1.0097163349 570 / 578 The train loss 1.00845199908 571 / 578 The train loss 1.0071522209 572 / 578 The train loss 1.00600704135 573 / 578 The train loss 1.00581721185 574 / 578 The train loss 1.00559667948 575 / 578 The train loss 1.00472088618 576 / 578 The train loss 1.00414753003 577 / 578 The train loss 1.00363094163 578 / 578 The train loss 1.00301876913 Starting epoch 2 Validation: 1 / 30 The valid loss 0.461336702108 2 / 30 The valid loss 0.435881480575 3 / 30 The valid loss 0.4136886398 4 / 30 The valid loss 0.390344411135 5 / 30 The valid loss 0.370161664486 6 / 30 The valid loss 0.409936904907 7 / 30 The valid loss 0.524121999741 8 / 30 The valid loss 0.669056862593 9 / 30 The valid loss 0.817215416167 10 / 30 The valid loss 0.897582161427 11 / 30 The valid loss 0.915576425466 12 / 30 The valid loss 0.926673263311 13 / 30 The valid loss 0.935562106279 14 / 30 The valid loss 0.940325660365 15 / 30 The valid loss 0.923230838776 16 / 30 The valid loss 0.915900740772 17 / 30 The valid loss 0.928968776675 18 / 30 The valid loss 0.938702990611 19 / 30 The valid loss 0.939050313674 20 / 30 The valid loss 0.932860502601 21 / 30 The valid loss 0.91891596147 22 / 30 The valid loss 0.905958335508 23 / 30 The valid loss 0.894222599009 24 / 30 The valid loss 0.885527436932 25 / 30 The valid loss 0.87723025322 26 / 30 The valid loss 0.873675699417 27 / 30 The valid loss 0.86287424741 28 / 30 The valid loss 0.84337372865 29 / 30 The valid loss 0.821272382962 30 / 30 The valid loss 0.802814881504 Training 1 / 578 The train loss 0.558399796486 2 / 578 The train loss 0.418712854385 3 / 578 The train loss 0.366999367873 4 / 578 The train loss 0.351666674018 5 / 578 The train loss 0.353314733505 6 / 578 The train loss 0.326589487493 7 / 578 The train loss 0.306191142116 8 / 578 The train loss 0.285148799419 9 / 578 The train loss 0.264974661171 10 / 578 The train loss 0.248987425119 11 / 578 The train loss 0.237382690338 12 / 578 The train loss 0.228293043251 13 / 578 The train loss 0.227047268588 14 / 578 The train loss 0.235671762377 15 / 578 The train loss 0.235612902542 16 / 578 The train loss 0.236348654609 17 / 578 The train loss 0.236073042978 18 / 578 The train loss 0.230235346076 19 / 578 The train loss 0.222131987936 20 / 578 The train loss 0.216153321788 21 / 578 The train loss 0.210352695059 22 / 578 The train loss 0.204110494053 23 / 578 The train loss 0.198330915816 24 / 578 The train loss 0.193926135389 25 / 578 The train loss 0.190317718685 26 / 578 The train loss 0.187329408354 27 / 578 The train loss 0.184907226375 28 / 578 The train loss 0.183114247397 29 / 578 The train loss 0.18211678878 30 / 578 The train loss 0.18081832851 31 / 578 The train loss 0.179189808667 32 / 578 The train loss 0.177907507168 33 / 578 The train loss 0.177620520872 34 / 578 The train loss 0.178107526153 35 / 578 The train loss 0.180562661162 36 / 578 The train loss 0.18475609252 37 / 578 The train loss 0.189101297912 38 / 578 The train loss 0.19202723609 39 / 578 The train loss 0.194857500876 40 / 578 The train loss 0.195336298086 41 / 578 The train loss 0.194339797809 42 / 578 The train loss 0.194550953983 43 / 578 The train loss 0.195772606446 44 / 578 The train loss 0.197478291833 45 / 578 The train loss 0.198671897418 46 / 578 The train loss 0.198185191692 47 / 578 The train loss 0.197031261598 48 / 578 The train loss 0.19509179409 49 / 578 The train loss 0.194621349628 50 / 578 The train loss 0.196189802438 51 / 578 The train loss 0.198354916391 52 / 578 The train loss 0.201794657713 53 / 578 The train loss 0.203563895006 54 / 578 The train loss 0.203539409433 55 / 578 The train loss 0.202953596142 56 / 578 The train loss 0.202556957225 57 / 578 The train loss 0.203445186348 58 / 578 The train loss 0.206527813884 59 / 578 The train loss 0.212467248283 60 / 578 The train loss 0.219903799022 61 / 578 The train loss 0.230489833555 62 / 578 The train loss 0.240270563792 63 / 578 The train loss 0.247651172181 64 / 578 The train loss 0.254720282857 65 / 578 The train loss 0.258422708167 66 / 578 The train loss 0.259922391199 67 / 578 The train loss 0.261355260081 68 / 578 The train loss 0.262046592748 69 / 578 The train loss 0.26328550873 70 / 578 The train loss 0.266309702077 71 / 578 The train loss 0.271238434168 72 / 578 The train loss 0.277312217177 73 / 578 The train loss 0.281841125901 74 / 578 The train loss 0.285155364488 75 / 578 The train loss 0.28978883634 76 / 578 The train loss 0.292612786062 77 / 578 The train loss 0.294519297972 78 / 578 The train loss 0.297258624377 79 / 578 The train loss 0.301816681989 80 / 578 The train loss 0.303602646757 81 / 578 The train loss 0.304713765872 82 / 578 The train loss 0.304950392192 83 / 578 The train loss 0.304445895176 84 / 578 The train loss 0.305447069041 85 / 578 The train loss 0.306102792743 86 / 578 The train loss 0.305595723436 87 / 578 The train loss 0.305092540024 88 / 578 The train loss 0.303650896065 89 / 578 The train loss 0.304267441373 90 / 578 The train loss 0.304997255156 91 / 578 The train loss 0.303472042493 92 / 578 The train loss 0.303618387117 93 / 578 The train loss 0.303366858873 94 / 578 The train loss 0.301782582431 95 / 578 The train loss 0.300650347612 96 / 578 The train loss 0.298957133743 97 / 578 The train loss 0.297803109685 98 / 578 The train loss 0.297028964956 99 / 578 The train loss 0.29736113315 100 / 578 The train loss 0.298216663077 101 / 578 The train loss 0.299906538338 102 / 578 The train loss 0.30091483314 103 / 578 The train loss 0.300978988481 104 / 578 The train loss 0.301315823785 105 / 578 The train loss 0.302327642909 106 / 578 The train loss 0.302653550047 107 / 578 The train loss 0.303574714775 108 / 578 The train loss 0.304815631843 109 / 578 The train loss 0.304597199032 110 / 578 The train loss 0.304572571204 111 / 578 The train loss 0.305621015663 112 / 578 The train loss 0.307249283218 113 / 578 The train loss 0.307955533332 114 / 578 The train loss 0.310349997002 115 / 578 The train loss 0.312382663139 116 / 578 The train loss 0.314606910957 117 / 578 The train loss 0.316667271284 118 / 578 The train loss 0.320250408945 119 / 578 The train loss 0.324130020535 120 / 578 The train loss 0.328380768312 121 / 578 The train loss 0.333321321417 122 / 578 The train loss 0.33944781033 123 / 578 The train loss 0.347331389482 124 / 578 The train loss 0.357930227934 125 / 578 The train loss 0.368646955788 126 / 578 The train loss 0.377966906639 127 / 578 The train loss 0.388195651077 128 / 578 The train loss 0.399375931651 129 / 578 The train loss 0.410742982702 130 / 578 The train loss 0.423472664276 131 / 578 The train loss 0.433024798015 132 / 578 The train loss 0.43761580127 133 / 578 The train loss 0.439827824278 134 / 578 The train loss 0.441625532208 135 / 578 The train loss 0.445155094564 136 / 578 The train loss 0.449946636937 137 / 578 The train loss 0.455080264948 138 / 578 The train loss 0.458043258529 139 / 578 The train loss 0.460433024993 140 / 578 The train loss 0.461265152959 141 / 578 The train loss 0.461101668999 142 / 578 The train loss 0.461621429351 143 / 578 The train loss 0.463345739027 144 / 578 The train loss 0.465418568554 145 / 578 The train loss 0.467190920125 146 / 578 The train loss 0.468277735712 147 / 578 The train loss 0.468150807735 148 / 578 The train loss 0.467680303279 149 / 578 The train loss 0.466863701058 150 / 578 The train loss 0.467429949492 151 / 578 The train loss 0.465873101837 152 / 578 The train loss 0.464537963026 153 / 578 The train loss 0.46334506284 154 / 578 The train loss 0.46242469196 155 / 578 The train loss 0.461958652879 156 / 578 The train loss 0.462270657938 157 / 578 The train loss 0.46337973763 158 / 578 The train loss 0.464307356814 159 / 578 The train loss 0.464830333285 160 / 578 The train loss 0.46500689066 161 / 578 The train loss 0.464903874196 162 / 578 The train loss 0.464772736723 163 / 578 The train loss 0.464572037427 164 / 578 The train loss 0.464037403782 165 / 578 The train loss 0.463787068336 166 / 578 The train loss 0.46424246943 167 / 578 The train loss 0.46360984226 168 / 578 The train loss 0.466522688445 169 / 578 The train loss 0.472362396765 170 / 578 The train loss 0.479389224815 171 / 578 The train loss 0.48604889689 172 / 578 The train loss 0.491593007908 173 / 578 The train loss 0.49870236808 174 / 578 The train loss 0.505110902134 175 / 578 The train loss 0.512258161604 176 / 578 The train loss 0.519194281025 177 / 578 The train loss 0.523506641851 178 / 578 The train loss 0.526663960826 179 / 578 The train loss 0.53104111315 180 / 578 The train loss 0.534748077186 181 / 578 The train loss 0.536537721465 182 / 578 The train loss 0.537440334498 183 / 578 The train loss 0.538782826986 184 / 578 The train loss 0.543359924067 185 / 578 The train loss 0.54936481082 186 / 578 The train loss 0.560190251237 187 / 578 The train loss 0.57615492222 188 / 578 The train loss 0.591020382901 189 / 578 The train loss 0.603169915971 190 / 578 The train loss 0.620216398921 191 / 578 The train loss 0.633823344026 192 / 578 The train loss 0.645815514455 193 / 578 The train loss 0.656387917543 194 / 578 The train loss 0.664195083274 195 / 578 The train loss 0.669005346719 196 / 578 The train loss 0.673541274416 197 / 578 The train loss 0.681768138813 198 / 578 The train loss 0.685643656499 199 / 578 The train loss 0.690173667093 200 / 578 The train loss 0.694944893532 201 / 578 The train loss 0.704879374461 202 / 578 The train loss 0.718328184826 203 / 578 The train loss 0.730600869531 204 / 578 The train loss 0.738832670501 205 / 578 The train loss 0.744857618685 206 / 578 The train loss 0.74752481799 207 / 578 The train loss 0.751137576061 208 / 578 The train loss 0.759106187114 209 / 578 The train loss 0.770393675416 210 / 578 The train loss 0.778190512139 211 / 578 The train loss 0.784194072099 212 / 578 The train loss 0.791234961482 213 / 578 The train loss 0.797285690379 214 / 578 The train loss 0.802022182879 215 / 578 The train loss 0.805911004647 216 / 578 The train loss 0.809057505422 217 / 578 The train loss 0.81057122209 218 / 578 The train loss 0.812427907738 219 / 578 The train loss 0.815118487397 220 / 578 The train loss 0.818091259592 221 / 578 The train loss 0.821851668686 222 / 578 The train loss 0.825554965685 223 / 578 The train loss 0.829952475034 224 / 578 The train loss 0.835151168218 225 / 578 The train loss 0.841031916241 226 / 578 The train loss 0.848361498849 227 / 578 The train loss 0.855710145496 228 / 578 The train loss 0.862285235831 229 / 578 The train loss 0.868140245987 230 / 578 The train loss 0.874681151788 231 / 578 The train loss 0.880315805765 232 / 578 The train loss 0.885908277224 233 / 578 The train loss 0.892370328723 234 / 578 The train loss 0.898692569512 235 / 578 The train loss 0.906906359595 236 / 578 The train loss 0.912203114259 237 / 578 The train loss 0.913164800575 238 / 578 The train loss 0.912791910386 239 / 578 The train loss 0.91272489605 240 / 578 The train loss 0.912975668007 241 / 578 The train loss 0.914101958368 242 / 578 The train loss 0.916083708867 243 / 578 The train loss 0.918237125524 244 / 578 The train loss 0.921757193465 245 / 578 The train loss 0.927213767688 246 / 578 The train loss 0.931739736954 247 / 578 The train loss 0.933580563226 248 / 578 The train loss 0.934859913774 249 / 578 The train loss 0.936379379148 250 / 578 The train loss 0.94030096063 251 / 578 The train loss 0.945134628431 252 / 578 The train loss 0.949859522727 253 / 578 The train loss 0.956568184666 254 / 578 The train loss 0.965662685993 255 / 578 The train loss 0.976496096306 256 / 578 The train loss 0.988695502369 257 / 578 The train loss 1.00000065133 258 / 578 The train loss 1.00947171967 259 / 578 The train loss 1.01463779896 260 / 578 The train loss 1.02008750264 261 / 578 The train loss 1.02535886974 262 / 578 The train loss 1.02971019644 263 / 578 The train loss 1.03460025615 264 / 578 The train loss 1.04058764929 265 / 578 The train loss 1.04626929543 266 / 578 The train loss 1.05210167492 267 / 578 The train loss 1.05846071341 268 / 578 The train loss 1.06701237327 269 / 578 The train loss 1.07753896633 270 / 578 The train loss 1.0891568847 271 / 578 The train loss 1.09859770699 272 / 578 The train loss 1.10745863362 273 / 578 The train loss 1.11663113679 274 / 578 The train loss 1.12585261364 275 / 578 The train loss 1.13496354285 276 / 578 The train loss 1.14264073933 277 / 578 The train loss 1.15043427806 278 / 578 The train loss 1.15861263986 279 / 578 The train loss 1.16671781736 280 / 578 The train loss 1.17628145992 281 / 578 The train loss 1.18590238773 282 / 578 The train loss 1.19459314473 283 / 578 The train loss 1.20405772726 284 / 578 The train loss 1.21242361345 285 / 578 The train loss 1.22015631834 286 / 578 The train loss 1.22790149883 287 / 578 The train loss 1.23502703336 288 / 578 The train loss 1.24220156926 289 / 578 The train loss 1.24936872625 290 / 578 The train loss 1.25732813711 291 / 578 The train loss 1.26482652479 292 / 578 The train loss 1.27429931057 293 / 578 The train loss 1.28232892226 294 / 578 The train loss 1.29041035653 295 / 578 The train loss 1.29753432443 296 / 578 The train loss 1.30433934854 297 / 578 The train loss 1.30993752937 298 / 578 The train loss 1.31552256921 299 / 578 The train loss 1.32220687363 300 / 578 The train loss 1.32945075678 301 / 578 The train loss 1.33491705192 302 / 578 The train loss 1.34172750711 303 / 578 The train loss 1.34893118595 304 / 578 The train loss 1.35434652402 305 / 578 The train loss 1.35925214572 306 / 578 The train loss 1.36330323367 307 / 578 The train loss 1.3677818706 308 / 578 The train loss 1.37229574585 309 / 578 The train loss 1.37639552883 310 / 578 The train loss 1.37737214765 311 / 578 The train loss 1.37678808039 312 / 578 The train loss 1.37533247574 313 / 578 The train loss 1.37326397937 314 / 578 The train loss 1.37156437772 315 / 578 The train loss 1.37039628528 316 / 578 The train loss 1.36969163606 317 / 578 The train loss 1.36963251231 318 / 578 The train loss 1.36928120544 319 / 578 The train loss 1.36800406278 320 / 578 The train loss 1.36627483859 321 / 578 The train loss 1.36332254031 322 / 578 The train loss 1.35972860661 323 / 578 The train loss 1.35628880927 324 / 578 The train loss 1.3527658363 325 / 578 The train loss 1.34921782638 326 / 578 The train loss 1.34531647321 327 / 578 The train loss 1.34153398794 328 / 578 The train loss 1.33778481826 329 / 578 The train loss 1.33422188683 330 / 578 The train loss 1.33055415853 331 / 578 The train loss 1.32701129716 332 / 578 The train loss 1.32343219728 333 / 578 The train loss 1.32035205104 334 / 578 The train loss 1.31760152661 335 / 578 The train loss 1.31645083379 336 / 578 The train loss 1.31601208009 337 / 578 The train loss 1.31642654339 338 / 578 The train loss 1.31695991177 339 / 578 The train loss 1.31666802362 340 / 578 The train loss 1.31597545519 341 / 578 The train loss 1.31345231342 342 / 578 The train loss 1.31081673316 343 / 578 The train loss 1.30841108822 344 / 578 The train loss 1.30607388935 345 / 578 The train loss 1.30397614176 346 / 578 The train loss 1.30199848301 347 / 578 The train loss 1.30028823838 348 / 578 The train loss 1.29918492239 349 / 578 The train loss 1.29795411361 350 / 578 The train loss 1.29725249533 351 / 578 The train loss 1.29672265057 352 / 578 The train loss 1.29594138574 353 / 578 The train loss 1.29476648455 354 / 578 The train loss 1.2936818196 355 / 578 The train loss 1.29292721215 356 / 578 The train loss 1.29260694114 357 / 578 The train loss 1.29262278298 358 / 578 The train loss 1.29218072003 359 / 578 The train loss 1.29122043142 360 / 578 The train loss 1.29009950215 361 / 578 The train loss 1.28964058799 362 / 578 The train loss 1.28841071749 363 / 578 The train loss 1.28671614793 364 / 578 The train loss 1.28487663728 365 / 578 The train loss 1.2835739982 366 / 578 The train loss 1.28031696801 367 / 578 The train loss 1.27702744841 368 / 578 The train loss 1.2738372768 369 / 578 The train loss 1.27063568364 370 / 578 The train loss 1.26738598991 371 / 578 The train loss 1.26420315952 372 / 578 The train loss 1.26109340152 373 / 578 The train loss 1.25796920539 374 / 578 The train loss 1.25490922433 375 / 578 The train loss 1.2519362721 376 / 578 The train loss 1.24896677659 377 / 578 The train loss 1.24595794352 378 / 578 The train loss 1.24293763155 379 / 578 The train loss 1.23994283005 380 / 578 The train loss 1.237006018 381 / 578 The train loss 1.23443495457 382 / 578 The train loss 1.23213676457 383 / 578 The train loss 1.2297486422 384 / 578 The train loss 1.22735577764 385 / 578 The train loss 1.2247734698 386 / 578 The train loss 1.22197386387 387 / 578 The train loss 1.21922596484 388 / 578 The train loss 1.21653138766 389 / 578 The train loss 1.21389476603 390 / 578 The train loss 1.21116188042 391 / 578 The train loss 1.20836400401 392 / 578 The train loss 1.2056021377 393 / 578 The train loss 1.20279547559 394 / 578 The train loss 1.200050251 395 / 578 The train loss 1.19723392021 396 / 578 The train loss 1.19483692739 397 / 578 The train loss 1.19289231434 398 / 578 The train loss 1.19108984778 399 / 578 The train loss 1.18920530052 400 / 578 The train loss 1.1872985225 401 / 578 The train loss 1.18537344627 402 / 578 The train loss 1.1835678376 403 / 578 The train loss 1.18218575257 404 / 578 The train loss 1.18099660513 405 / 578 The train loss 1.1800083573 406 / 578 The train loss 1.17903486986 407 / 578 The train loss 1.17782880639 408 / 578 The train loss 1.17660010422 409 / 578 The train loss 1.17785956725 410 / 578 The train loss 1.18109406355 411 / 578 The train loss 1.18339018961 412 / 578 The train loss 1.18451522397 413 / 578 The train loss 1.18528236811 414 / 578 The train loss 1.18675047136 415 / 578 The train loss 1.18767624797 416 / 578 The train loss 1.18747411416 417 / 578 The train loss 1.18694426364 418 / 578 The train loss 1.18668184602 419 / 578 The train loss 1.18641632943 420 / 578 The train loss 1.18646252601 421 / 578 The train loss 1.18643112044 422 / 578 The train loss 1.18627818718 423 / 578 The train loss 1.18646121524 424 / 578 The train loss 1.18636702133 425 / 578 The train loss 1.18611168123 426 / 578 The train loss 1.18560965805 427 / 578 The train loss 1.18451971407 428 / 578 The train loss 1.18329550874 429 / 578 The train loss 1.18187685979 430 / 578 The train loss 1.17995884458 431 / 578 The train loss 1.17789138641 432 / 578 The train loss 1.17594524706 433 / 578 The train loss 1.17374128463 434 / 578 The train loss 1.17132319527 435 / 578 The train loss 1.16893936865 436 / 578 The train loss 1.16663545261 437 / 578 The train loss 1.16479218136 438 / 578 The train loss 1.16339466119 439 / 578 The train loss 1.16234409205 440 / 578 The train loss 1.16110609221 441 / 578 The train loss 1.15985952706 442 / 578 The train loss 1.15856751132 443 / 578 The train loss 1.15746655549 444 / 578 The train loss 1.15714269279 445 / 578 The train loss 1.15714732763 446 / 578 The train loss 1.15776011755 447 / 578 The train loss 1.15808532375 448 / 578 The train loss 1.15843607425 449 / 578 The train loss 1.15831070773 450 / 578 The train loss 1.15816294531 451 / 578 The train loss 1.15797623688 452 / 578 The train loss 1.15782543087 453 / 578 The train loss 1.1575723199 454 / 578 The train loss 1.15737071272 455 / 578 The train loss 1.15701966318 456 / 578 The train loss 1.1563271541 457 / 578 The train loss 1.15574424044 458 / 578 The train loss 1.15488877417 459 / 578 The train loss 1.15412820392 460 / 578 The train loss 1.15362833369 461 / 578 The train loss 1.15295382339 462 / 578 The train loss 1.15211302087 463 / 578 The train loss 1.1509164481 464 / 578 The train loss 1.14879700233 465 / 578 The train loss 1.14662180037 466 / 578 The train loss 1.14445289905 467 / 578 The train loss 1.14227631519 468 / 578 The train loss 1.14028692252 469 / 578 The train loss 1.13829890955 470 / 578 The train loss 1.13640300381 471 / 578 The train loss 1.13453622118 472 / 578 The train loss 1.13262037526 473 / 578 The train loss 1.13066822602 474 / 578 The train loss 1.12863328943 475 / 578 The train loss 1.12665896275 476 / 578 The train loss 1.12472902041 477 / 578 The train loss 1.12279062973 478 / 578 The train loss 1.12064350538 479 / 578 The train loss 1.11849891705 480 / 578 The train loss 1.1163949296 481 / 578 The train loss 1.11438459317 482 / 578 The train loss 1.11232794087 483 / 578 The train loss 1.11024655863 484 / 578 The train loss 1.10819587542 485 / 578 The train loss 1.10621046021 486 / 578 The train loss 1.10425588216 487 / 578 The train loss 1.10229338167 488 / 578 The train loss 1.10035995476 489 / 578 The train loss 1.09842331717 490 / 578 The train loss 1.09642019527 491 / 578 The train loss 1.09457140912 492 / 578 The train loss 1.09269793953 493 / 578 The train loss 1.09086718738 494 / 578 The train loss 1.08894455418 495 / 578 The train loss 1.08702855736 496 / 578 The train loss 1.08548405646 497 / 578 The train loss 1.08410647709 498 / 578 The train loss 1.0826052018 499 / 578 The train loss 1.08103745465 500 / 578 The train loss 1.07934940031 501 / 578 The train loss 1.07753155031 502 / 578 The train loss 1.07582550905 503 / 578 The train loss 1.07410857775 504 / 578 The train loss 1.07241580026 505 / 578 The train loss 1.07080444428 506 / 578 The train loss 1.06902293549 507 / 578 The train loss 1.06733957959 508 / 578 The train loss 1.06553852543 509 / 578 The train loss 1.06370521704 510 / 578 The train loss 1.06182965499 511 / 578 The train loss 1.05991846227 512 / 578 The train loss 1.05808546713 513 / 578 The train loss 1.05620458689 514 / 578 The train loss 1.05440996667 515 / 578 The train loss 1.05265330781 516 / 578 The train loss 1.05093462357 517 / 578 The train loss 1.04924504181 518 / 578 The train loss 1.04754151906 519 / 578 The train loss 1.04586506362 520 / 578 The train loss 1.04423210379 521 / 578 The train loss 1.04265040358 522 / 578 The train loss 1.04095272002 523 / 578 The train loss 1.03927172807 524 / 578 The train loss 1.03782134564 525 / 578 The train loss 1.03632002628 526 / 578 The train loss 1.03481905096 527 / 578 The train loss 1.03347780861 528 / 578 The train loss 1.03219227031 529 / 578 The train loss 1.0310779289 530 / 578 The train loss 1.02993694017 531 / 578 The train loss 1.02857306017 532 / 578 The train loss 1.02724851428 533 / 578 The train loss 1.02594002775 534 / 578 The train loss 1.02442479163 535 / 578 The train loss 1.02282419663 536 / 578 The train loss 1.0212595724 537 / 578 The train loss 1.01972911335 538 / 578 The train loss 1.01835336623 539 / 578 The train loss 1.01698750624 540 / 578 The train loss 1.01559224881 541 / 578 The train loss 1.0143220135 542 / 578 The train loss 1.01308614044 543 / 578 The train loss 1.01161585192 544 / 578 The train loss 1.01030731833 545 / 578 The train loss 1.00937459913 546 / 578 The train loss 1.00952907454 547 / 578 The train loss 1.01009332182 548 / 578 The train loss 1.01228178751 549 / 578 The train loss 1.01336506445 550 / 578 The train loss 1.01375477017 551 / 578 The train loss 1.01385664146 552 / 578 The train loss 1.01383528379 553 / 578 The train loss 1.01381421225 554 / 578 The train loss 1.01374288174 555 / 578 The train loss 1.01335532189 556 / 578 The train loss 1.01290314464 557 / 578 The train loss 1.01258843737 558 / 578 The train loss 1.01218341923 559 / 578 The train loss 1.01188076562 560 / 578 The train loss 1.01167729929 561 / 578 The train loss 1.0114710814 562 / 578 The train loss 1.01136855517 563 / 578 The train loss 1.01083021103 564 / 578 The train loss 1.00951952668 565 / 578 The train loss 1.00820760649 566 / 578 The train loss 1.00695345597 567 / 578 The train loss 1.00566487638 568 / 578 The train loss 1.00438606048 569 / 578 The train loss 1.00317438244 570 / 578 The train loss 1.0018955143 571 / 578 The train loss 1.00056694217 572 / 578 The train loss 0.999437198733 573 / 578 The train loss 0.999288154691 574 / 578 The train loss 0.998949831021 575 / 578 The train loss 0.998135726413 576 / 578 The train loss 0.997545235752 577 / 578 The train loss 0.996879726661 578 / 578 The train loss 0.996237571535 Starting epoch 3 Validation: 1 / 30 The valid loss 0.460784345865 2 / 30 The valid loss 0.435379177332 3 / 30 The valid loss 0.413092037042 4 / 30 The valid loss 0.389243915677 5 / 30 The valid loss 0.368766969442 6 / 30 The valid loss 0.408340786894 7 / 30 The valid loss 0.522320044892 8 / 30 The valid loss 0.667099159211 9 / 30 The valid loss 0.815135594871 10 / 30 The valid loss 0.895409092307 11 / 30 The valid loss 0.91331414472 12 / 30 The valid loss 0.924322215219 13 / 30 The valid loss 0.933136616762 14 / 30 The valid loss 0.93784418915 15 / 30 The valid loss 0.920772598187 16 / 30 The valid loss 0.913457227871 17 / 30 The valid loss 0.926504596191 18 / 30 The valid loss 0.936211067769 19 / 30 The valid loss 0.93654053619 20 / 30 The valid loss 0.930362452567 21 / 30 The valid loss 0.91641057531 22 / 30 The valid loss 0.90345176648 23 / 30 The valid loss 0.891725046479 24 / 30 The valid loss 0.883043119063 25 / 30 The valid loss 0.874727605581 26 / 30 The valid loss 0.87115476567 27 / 30 The valid loss 0.860347996155 28 / 30 The valid loss 0.840866930783 29 / 30 The valid loss 0.818836713146 30 / 30 The valid loss 0.8004650111 Training 1 / 578 The train loss 0.614065885544 2 / 578 The train loss 0.476967617869 3 / 578 The train loss 0.415765513976 4 / 578 The train loss 0.399163730443 5 / 578 The train loss 0.383987414837 6 / 578 The train loss 0.35549774766 7 / 578 The train loss 0.331041719232 8 / 578 The train loss 0.308947680518 9 / 578 The train loss 0.283599363433 10 / 578 The train loss 0.265755201131 11 / 578 The train loss 0.251788311384 12 / 578 The train loss 0.242237719397 13 / 578 The train loss 0.24025823749 14 / 578 The train loss 0.250552926745 15 / 578 The train loss 0.250160469611 16 / 578 The train loss 0.248722155578 17 / 578 The train loss 0.246597067398 18 / 578 The train loss 0.241028394136 19 / 578 The train loss 0.233169952506 20 / 578 The train loss 0.225467022881 21 / 578 The train loss 0.219148678794 22 / 578 The train loss 0.213323353028 23 / 578 The train loss 0.206918138849 24 / 578 The train loss 0.201768238408 25 / 578 The train loss 0.197846720219 26 / 578 The train loss 0.193780043377 27 / 578 The train loss 0.190702177032 28 / 578 The train loss 0.190020251753 29 / 578 The train loss 0.18766199746 30 / 578 The train loss 0.185453070452 31 / 578 The train loss 0.184135371879 32 / 578 The train loss 0.182501239935 33 / 578 The train loss 0.182071424569 34 / 578 The train loss 0.181629082955 35 / 578 The train loss 0.183626706472 36 / 578 The train loss 0.188804254971 37 / 578 The train loss 0.192396355843 38 / 578 The train loss 0.196234547974 39 / 578 The train loss 0.19885657785 40 / 578 The train loss 0.199245774932 41 / 578 The train loss 0.198731678833 42 / 578 The train loss 0.198325996065 43 / 578 The train loss 0.200369747052 44 / 578 The train loss 0.20174545473 45 / 578 The train loss 0.201784265869 46 / 578 The train loss 0.201246832538 47 / 578 The train loss 0.199988799685 48 / 578 The train loss 0.198833692043 49 / 578 The train loss 0.19827601268 50 / 578 The train loss 0.199726875573 51 / 578 The train loss 0.201854070758 52 / 578 The train loss 0.203492381968 53 / 578 The train loss 0.205316673051 54 / 578 The train loss 0.205834687998 55 / 578 The train loss 0.206061345474 56 / 578 The train loss 0.205665258558 57 / 578 The train loss 0.208047471679 58 / 578 The train loss 0.210510137395 59 / 578 The train loss 0.215123169629 60 / 578 The train loss 0.222657546525 61 / 578 The train loss 0.231936108016 62 / 578 The train loss 0.24265329023 63 / 578 The train loss 0.250900905402 64 / 578 The train loss 0.256956911762 65 / 578 The train loss 0.261441518825 66 / 578 The train loss 0.262745701115 67 / 578 The train loss 0.264490897642 68 / 578 The train loss 0.265307185076 69 / 578 The train loss 0.267504804916 70 / 578 The train loss 0.270725204796 71 / 578 The train loss 0.275247933684 72 / 578 The train loss 0.281438201355 73 / 578 The train loss 0.286593643464 74 / 578 The train loss 0.289959570544 75 / 578 The train loss 0.295233533879 76 / 578 The train loss 0.297087038132 77 / 578 The train loss 0.297961751072 78 / 578 The train loss 0.300335040937 79 / 578 The train loss 0.30263920383 80 / 578 The train loss 0.30436828332 81 / 578 The train loss 0.30476193452 82 / 578 The train loss 0.304374274412 83 / 578 The train loss 0.304726614381 84 / 578 The train loss 0.304825896752 85 / 578 The train loss 0.304254216043 86 / 578 The train loss 0.30303326483 87 / 578 The train loss 0.301752749103 88 / 578 The train loss 0.300465503962 89 / 578 The train loss 0.301748309326 90 / 578 The train loss 0.301837408625 91 / 578 The train loss 0.300489790083 92 / 578 The train loss 0.299201052231 93 / 578 The train loss 0.298251189291 94 / 578 The train loss 0.296591506597 95 / 578 The train loss 0.294618247136 96 / 578 The train loss 0.293719963869 97 / 578 The train loss 0.292828062069 98 / 578 The train loss 0.292900338085 99 / 578 The train loss 0.293090826258 100 / 578 The train loss 0.293728661165 101 / 578 The train loss 0.294658549868 102 / 578 The train loss 0.29561593683 103 / 578 The train loss 0.295361629317 104 / 578 The train loss 0.295381637362 105 / 578 The train loss 0.295128427162 106 / 578 The train loss 0.295315816647 107 / 578 The train loss 0.295631485495 108 / 578 The train loss 0.295494644385 109 / 578 The train loss 0.295384775266 110 / 578 The train loss 0.295414047553 111 / 578 The train loss 0.297104064073 112 / 578 The train loss 0.29734034803 113 / 578 The train loss 0.297855558583 114 / 578 The train loss 0.3010874986 115 / 578 The train loss 0.303302286436 116 / 578 The train loss 0.306342575506 117 / 578 The train loss 0.309714782888 118 / 578 The train loss 0.313028063246 119 / 578 The train loss 0.316074636741 120 / 578 The train loss 0.320164307393 121 / 578 The train loss 0.323888598826 122 / 578 The train loss 0.328147595473 123 / 578 The train loss 0.33621165853 124 / 578 The train loss 0.346618328784 125 / 578 The train loss 0.35648457092 126 / 578 The train loss 0.366140475763 127 / 578 The train loss 0.377661435095 128 / 578 The train loss 0.387370251527 129 / 578 The train loss 0.398448763372 130 / 578 The train loss 0.411060867344 131 / 578 The train loss 0.42090650227 132 / 578 The train loss 0.425487572892 133 / 578 The train loss 0.428028197609 134 / 578 The train loss 0.429770802773 135 / 578 The train loss 0.433003847191 136 / 578 The train loss 0.437705249854 137 / 578 The train loss 0.4432253911 138 / 578 The train loss 0.446123758466 139 / 578 The train loss 0.448040357543 140 / 578 The train loss 0.448876552443 141 / 578 The train loss 0.448466110684 142 / 578 The train loss 0.448553120744 143 / 578 The train loss 0.450397105223 144 / 578 The train loss 0.451991128166 145 / 578 The train loss 0.453155828087 146 / 578 The train loss 0.453497248715 147 / 578 The train loss 0.453494687216 148 / 578 The train loss 0.452586498162 149 / 578 The train loss 0.451922184619 150 / 578 The train loss 0.451553986818 151 / 578 The train loss 0.449645935453 152 / 578 The train loss 0.448030414697 153 / 578 The train loss 0.446031894024 154 / 578 The train loss 0.44449129154 155 / 578 The train loss 0.443596000585 156 / 578 The train loss 0.443575429563 157 / 578 The train loss 0.443563519722 158 / 578 The train loss 0.443411166108 159 / 578 The train loss 0.442915146034 160 / 578 The train loss 0.442259458592 161 / 578 The train loss 0.441126152919 162 / 578 The train loss 0.44037752272 163 / 578 The train loss 0.439702435825 164 / 578 The train loss 0.438692219692 165 / 578 The train loss 0.438028864743 166 / 578 The train loss 0.436820705029 167 / 578 The train loss 0.435466204798 168 / 578 The train loss 0.438998896673 169 / 578 The train loss 0.446692519154 170 / 578 The train loss 0.455205994012 171 / 578 The train loss 0.462913894035 172 / 578 The train loss 0.469457807453 173 / 578 The train loss 0.475324103813 174 / 578 The train loss 0.482164681857 175 / 578 The train loss 0.490928834038 176 / 578 The train loss 0.497729835397 177 / 578 The train loss 0.502780321838 178 / 578 The train loss 0.507170145641 179 / 578 The train loss 0.511547504399 180 / 578 The train loss 0.515074030765 181 / 578 The train loss 0.517694205416 182 / 578 The train loss 0.519325705225 183 / 578 The train loss 0.520785239428 184 / 578 The train loss 0.525817145674 185 / 578 The train loss 0.531382501649 186 / 578 The train loss 0.540756923096 187 / 578 The train loss 0.554984207659 188 / 578 The train loss 0.567190734828 189 / 578 The train loss 0.578590987418 190 / 578 The train loss 0.593595922817 191 / 578 The train loss 0.606570768302 192 / 578 The train loss 0.617902777041 193 / 578 The train loss 0.62724021081 194 / 578 The train loss 0.634246737264 195 / 578 The train loss 0.638643607918 196 / 578 The train loss 0.643478935052 197 / 578 The train loss 0.651429205566 198 / 578 The train loss 0.655439120539 199 / 578 The train loss 0.659570797791 200 / 578 The train loss 0.66528198678 201 / 578 The train loss 0.674504299847 202 / 578 The train loss 0.687484579105 203 / 578 The train loss 0.69951746811 204 / 578 The train loss 0.708733292954 205 / 578 The train loss 0.715686669764 206 / 578 The train loss 0.718980315422 207 / 578 The train loss 0.721937688832 208 / 578 The train loss 0.729568780794 209 / 578 The train loss 0.740989027019 210 / 578 The train loss 0.748542145498 211 / 578 The train loss 0.754473902214 212 / 578 The train loss 0.761446035309 213 / 578 The train loss 0.767714613655 214 / 578 The train loss 0.773165458448 215 / 578 The train loss 0.777203764091 216 / 578 The train loss 0.78013522002 217 / 578 The train loss 0.781633437126 218 / 578 The train loss 0.783103061003 219 / 578 The train loss 0.785420015148 220 / 578 The train loss 0.788014259291 221 / 578 The train loss 0.791673860028 222 / 578 The train loss 0.795825122344 223 / 578 The train loss 0.799767396008 224 / 578 The train loss 0.80440679901 225 / 578 The train loss 0.81126231369 226 / 578 The train loss 0.818731263313 227 / 578 The train loss 0.826397139234 228 / 578 The train loss 0.832505132709 229 / 578 The train loss 0.838767662482 230 / 578 The train loss 0.84476372849 231 / 578 The train loss 0.851421988533 232 / 578 The train loss 0.856838110866 233 / 578 The train loss 0.862743516497 234 / 578 The train loss 0.869573101688 235 / 578 The train loss 0.878940605261 236 / 578 The train loss 0.884835675264 237 / 578 The train loss 0.886330384903 238 / 578 The train loss 0.886239569634 239 / 578 The train loss 0.886473150589 240 / 578 The train loss 0.88727878 241 / 578 The train loss 0.889048892587 242 / 578 The train loss 0.89175532983 243 / 578 The train loss 0.89415936868 244 / 578 The train loss 0.897913412145 245 / 578 The train loss 0.903044197663 246 / 578 The train loss 0.908260892229 247 / 578 The train loss 0.910583894896 248 / 578 The train loss 0.912370392542 249 / 578 The train loss 0.914036174047 250 / 578 The train loss 0.918596166283 251 / 578 The train loss 0.923604779658 252 / 578 The train loss 0.928611855333 253 / 578 The train loss 0.935227125206 254 / 578 The train loss 0.944385747887 255 / 578 The train loss 0.955847556597 256 / 578 The train loss 0.969138335116 257 / 578 The train loss 0.980773597057 258 / 578 The train loss 0.990051889287 259 / 578 The train loss 0.996065989108 260 / 578 The train loss 1.00179619024 261 / 578 The train loss 1.00732909094 262 / 578 The train loss 1.01219048087 263 / 578 The train loss 1.01720772556 264 / 578 The train loss 1.02406634134 265 / 578 The train loss 1.03051964621 266 / 578 The train loss 1.03666959459 267 / 578 The train loss 1.04380562212 268 / 578 The train loss 1.05260606902 269 / 578 The train loss 1.06300234321 270 / 578 The train loss 1.07497739585 271 / 578 The train loss 1.08536706169 272 / 578 The train loss 1.09437041042 273 / 578 The train loss 1.10418581496 274 / 578 The train loss 1.11358023791 275 / 578 The train loss 1.12265153508 276 / 578 The train loss 1.13112011825 277 / 578 The train loss 1.13917091233 278 / 578 The train loss 1.1477456493 279 / 578 The train loss 1.1566263413 280 / 578 The train loss 1.16730227867 281 / 578 The train loss 1.17706128512 282 / 578 The train loss 1.18642535671 283 / 578 The train loss 1.19651861044 284 / 578 The train loss 1.2048977132 285 / 578 The train loss 1.21334854231 286 / 578 The train loss 1.22153058307 287 / 578 The train loss 1.22946154254 288 / 578 The train loss 1.23706320479 289 / 578 The train loss 1.24404100865 290 / 578 The train loss 1.25216537562 291 / 578 The train loss 1.26038124525 292 / 578 The train loss 1.26946638925 293 / 578 The train loss 1.27795291651 294 / 578 The train loss 1.286616377 295 / 578 The train loss 1.29470321498 296 / 578 The train loss 1.3013695822 297 / 578 The train loss 1.30808245024 298 / 578 The train loss 1.31423607981 299 / 578 The train loss 1.32120761158 300 / 578 The train loss 1.32890780501 301 / 578 The train loss 1.33616510313 302 / 578 The train loss 1.34397620843 303 / 578 The train loss 1.35080738953 304 / 578 The train loss 1.35716846166 305 / 578 The train loss 1.36252426371 306 / 578 The train loss 1.36785282937 307 / 578 The train loss 1.37308543486 308 / 578 The train loss 1.3778928655 309 / 578 The train loss 1.38188963543 310 / 578 The train loss 1.38354047318 311 / 578 The train loss 1.38344410045 312 / 578 The train loss 1.38244550706 313 / 578 The train loss 1.38065690423 314 / 578 The train loss 1.37912019378 315 / 578 The train loss 1.37786518253 316 / 578 The train loss 1.378090573 317 / 578 The train loss 1.37825968262 318 / 578 The train loss 1.3779634767 319 / 578 The train loss 1.37725235052 320 / 578 The train loss 1.3758984878 321 / 578 The train loss 1.37307940477 322 / 578 The train loss 1.36974810455 323 / 578 The train loss 1.36628883517 324 / 578 The train loss 1.36267977362 325 / 578 The train loss 1.35883925266 326 / 578 The train loss 1.35487038784 327 / 578 The train loss 1.35100263321 328 / 578 The train loss 1.34720430362 329 / 578 The train loss 1.34350415446 330 / 578 The train loss 1.33987968964 331 / 578 The train loss 1.3361486994 332 / 578 The train loss 1.33240924493 333 / 578 The train loss 1.32924660096 334 / 578 The train loss 1.32655628788 335 / 578 The train loss 1.32520998472 336 / 578 The train loss 1.32511829855 337 / 578 The train loss 1.32564892529 338 / 578 The train loss 1.32568149829 339 / 578 The train loss 1.32577163372 340 / 578 The train loss 1.3245826797 341 / 578 The train loss 1.322126037 342 / 578 The train loss 1.31929480032 343 / 578 The train loss 1.31696222662 344 / 578 The train loss 1.314619131 345 / 578 The train loss 1.31255348087 346 / 578 The train loss 1.31050440148 347 / 578 The train loss 1.30869492137 348 / 578 The train loss 1.30757471812 349 / 578 The train loss 1.30714547425 350 / 578 The train loss 1.30664917805 351 / 578 The train loss 1.30593986249 352 / 578 The train loss 1.30517434257 353 / 578 The train loss 1.30424995526 354 / 578 The train loss 1.30279679601 355 / 578 The train loss 1.30233146586 356 / 578 The train loss 1.30203319446 357 / 578 The train loss 1.30140408193 358 / 578 The train loss 1.30078773138 359 / 578 The train loss 1.29987616854 360 / 578 The train loss 1.29904112464 361 / 578 The train loss 1.29822571766 362 / 578 The train loss 1.29704077134 363 / 578 The train loss 1.29556823085 364 / 578 The train loss 1.2936503294 365 / 578 The train loss 1.29195111676 366 / 578 The train loss 1.28866425081 367 / 578 The train loss 1.28535294833 368 / 578 The train loss 1.28208512274 369 / 578 The train loss 1.27879707418 370 / 578 The train loss 1.2754633921 371 / 578 The train loss 1.27219129558 372 / 578 The train loss 1.26899918979 373 / 578 The train loss 1.265880112 374 / 578 The train loss 1.26278529371 375 / 578 The train loss 1.25969959568 376 / 578 The train loss 1.2565823527 377 / 578 The train loss 1.25349017526 378 / 578 The train loss 1.25042640405 379 / 578 The train loss 1.24745057575 380 / 578 The train loss 1.24445798292 381 / 578 The train loss 1.24177394824 382 / 578 The train loss 1.23935231166 383 / 578 The train loss 1.23692402754 384 / 578 The train loss 1.23435546455 385 / 578 The train loss 1.23180718806 386 / 578 The train loss 1.22901390853 387 / 578 The train loss 1.22615311829 388 / 578 The train loss 1.2233684267 389 / 578 The train loss 1.22074071482 390 / 578 The train loss 1.21802642664 391 / 578 The train loss 1.21517466172 392 / 578 The train loss 1.21235627739 393 / 578 The train loss 1.20959386958 394 / 578 The train loss 1.2068428218 395 / 578 The train loss 1.20399165654 396 / 578 The train loss 1.20149708538 397 / 578 The train loss 1.19935005677 398 / 578 The train loss 1.19745704194 399 / 578 The train loss 1.19551200443 400 / 578 The train loss 1.19359778977 401 / 578 The train loss 1.19162367629 402 / 578 The train loss 1.18977248549 403 / 578 The train loss 1.18841995316 404 / 578 The train loss 1.18702702209 405 / 578 The train loss 1.18581778958 406 / 578 The train loss 1.18463043625 407 / 578 The train loss 1.18343002174 408 / 578 The train loss 1.18230574139 409 / 578 The train loss 1.18381573736 410 / 578 The train loss 1.18696426458 411 / 578 The train loss 1.18905867262 412 / 578 The train loss 1.18995636976 413 / 578 The train loss 1.19076676763 414 / 578 The train loss 1.1920625517 415 / 578 The train loss 1.19317969858 416 / 578 The train loss 1.19281515894 417 / 578 The train loss 1.19209286291 418 / 578 The train loss 1.19142657004 419 / 578 The train loss 1.19129922139 420 / 578 The train loss 1.19126396093 421 / 578 The train loss 1.19113135422 422 / 578 The train loss 1.19090328057 423 / 578 The train loss 1.19088977814 424 / 578 The train loss 1.19089827447 425 / 578 The train loss 1.19033289894 426 / 578 The train loss 1.18946992245 427 / 578 The train loss 1.1883565514 428 / 578 The train loss 1.18705533152 429 / 578 The train loss 1.18559505111 430 / 578 The train loss 1.18355534007 431 / 578 The train loss 1.18153899845 432 / 578 The train loss 1.17950532516 433 / 578 The train loss 1.17718541955 434 / 578 The train loss 1.17473411631 435 / 578 The train loss 1.17228764024 436 / 578 The train loss 1.16990135421 437 / 578 The train loss 1.16800341756 438 / 578 The train loss 1.16659518113 439 / 578 The train loss 1.16542494476 440 / 578 The train loss 1.16422430401 441 / 578 The train loss 1.16285837997 442 / 578 The train loss 1.16164990628 443 / 578 The train loss 1.16038655134 444 / 578 The train loss 1.15983921896 445 / 578 The train loss 1.16000152132 446 / 578 The train loss 1.16060116649 447 / 578 The train loss 1.16121408326 448 / 578 The train loss 1.16137317637 449 / 578 The train loss 1.16127959885 450 / 578 The train loss 1.16108051475 451 / 578 The train loss 1.16094336504 452 / 578 The train loss 1.16074307081 453 / 578 The train loss 1.16033200215 454 / 578 The train loss 1.15998417299 455 / 578 The train loss 1.15932682414 456 / 578 The train loss 1.15885410134 457 / 578 The train loss 1.15827902574 458 / 578 The train loss 1.15752674934 459 / 578 The train loss 1.15686921385 460 / 578 The train loss 1.15658103956 461 / 578 The train loss 1.1560081408 462 / 578 The train loss 1.15503021987 463 / 578 The train loss 1.15385051978 464 / 578 The train loss 1.15173991412 465 / 578 The train loss 1.14960893216 466 / 578 The train loss 1.14746575661 467 / 578 The train loss 1.14528132406 468 / 578 The train loss 1.14315213874 469 / 578 The train loss 1.14115557341 470 / 578 The train loss 1.13918870921 471 / 578 The train loss 1.13731904185 472 / 578 The train loss 1.13543883 473 / 578 The train loss 1.13343818063 474 / 578 The train loss 1.13142170146 475 / 578 The train loss 1.12944547662 476 / 578 The train loss 1.12753914393 477 / 578 The train loss 1.12546376911 478 / 578 The train loss 1.12338861136 479 / 578 The train loss 1.12135517988 480 / 578 The train loss 1.11927563169 481 / 578 The train loss 1.11724094829 482 / 578 The train loss 1.11519453408 483 / 578 The train loss 1.11314449624 484 / 578 The train loss 1.11105538293 485 / 578 The train loss 1.10908508136 486 / 578 The train loss 1.1070290724 487 / 578 The train loss 1.1051070126 488 / 578 The train loss 1.10319906969 489 / 578 The train loss 1.10125011005 490 / 578 The train loss 1.09924149213 491 / 578 The train loss 1.09725047313 492 / 578 The train loss 1.09538109276 493 / 578 The train loss 1.09358124307 494 / 578 The train loss 1.09166232176 495 / 578 The train loss 1.08988029899 496 / 578 The train loss 1.08833878946 497 / 578 The train loss 1.08681288969 498 / 578 The train loss 1.08528138418 499 / 578 The train loss 1.08375031738 500 / 578 The train loss 1.08201321106 501 / 578 The train loss 1.08015076045 502 / 578 The train loss 1.07836364627 503 / 578 The train loss 1.07667161716 504 / 578 The train loss 1.07505953086 505 / 578 The train loss 1.07336842157 506 / 578 The train loss 1.07170306281 507 / 578 The train loss 1.069947447 508 / 578 The train loss 1.06807316483 509 / 578 The train loss 1.0661359279 510 / 578 The train loss 1.06420631014 511 / 578 The train loss 1.06232618008 512 / 578 The train loss 1.06044327681 513 / 578 The train loss 1.05854060091 514 / 578 The train loss 1.05670114544 515 / 578 The train loss 1.05491619415 516 / 578 The train loss 1.05325308804 517 / 578 The train loss 1.0515909608 518 / 578 The train loss 1.04985874515 519 / 578 The train loss 1.04810881601 520 / 578 The train loss 1.04650045729 521 / 578 The train loss 1.04484622925 522 / 578 The train loss 1.04307133467 523 / 578 The train loss 1.04134544408 524 / 578 The train loss 1.03988091985 525 / 578 The train loss 1.03840424693 526 / 578 The train loss 1.03692053391 527 / 578 The train loss 1.0355600845 528 / 578 The train loss 1.03433590881 529 / 578 The train loss 1.03319656271 530 / 578 The train loss 1.03197277809 531 / 578 The train loss 1.03068536236 532 / 578 The train loss 1.02933567624 533 / 578 The train loss 1.02804302707 534 / 578 The train loss 1.02657462832 535 / 578 The train loss 1.02500710208 536 / 578 The train loss 1.02339678357 537 / 578 The train loss 1.02186918508 538 / 578 The train loss 1.0205289543 539 / 578 The train loss 1.01914835346 540 / 578 The train loss 1.01776772141 541 / 578 The train loss 1.01647264376 542 / 578 The train loss 1.01528932038 543 / 578 The train loss 1.0138735424 544 / 578 The train loss 1.01274334088 545 / 578 The train loss 1.01184570064 546 / 578 The train loss 1.0120245469 547 / 578 The train loss 1.01291336391 548 / 578 The train loss 1.01528727483 549 / 578 The train loss 1.01632003985 550 / 578 The train loss 1.01638797171 551 / 578 The train loss 1.01652749066 552 / 578 The train loss 1.01637950115 553 / 578 The train loss 1.01630777139 554 / 578 The train loss 1.01622141566 555 / 578 The train loss 1.01588765347 556 / 578 The train loss 1.01533937794 557 / 578 The train loss 1.01499100013 558 / 578 The train loss 1.01468662962 559 / 578 The train loss 1.01429329145 560 / 578 The train loss 1.01410399842 561 / 578 The train loss 1.01374854407 562 / 578 The train loss 1.01350896593 563 / 578 The train loss 1.0130036582 564 / 578 The train loss 1.01173300192 565 / 578 The train loss 1.01043478089 566 / 578 The train loss 1.00914603463 567 / 578 The train loss 1.00784620145 568 / 578 The train loss 1.00652375987 569 / 578 The train loss 1.00519539331 570 / 578 The train loss 1.00395217576 571 / 578 The train loss 1.00258864509 572 / 578 The train loss 1.00144064949 573 / 578 The train loss 1.00135986532 574 / 578 The train loss 1.00101944598 575 / 578 The train loss 1.00004999795 576 / 578 The train loss 0.999411648055 577 / 578 The train loss 0.998836374178 578 / 578 The train loss 0.99821672392 Starting epoch 4 Validation: 1 / 30 The valid loss 0.461691737175 2 / 30 The valid loss 0.43730455637 3 / 30 The valid loss 0.41611991326 4 / 30 The valid loss 0.39541529119 5 / 30 The valid loss 0.376833504438 6 / 30 The valid loss 0.417710537712 7 / 30 The valid loss 0.532934797662 8 / 30 The valid loss 0.678643580526 9 / 30 The valid loss 0.827360202869 10 / 30 The valid loss 0.908146068454 11 / 30 The valid loss 0.926607400179 12 / 30 The valid loss 0.938200779259 13 / 30 The valid loss 0.947503646979 14 / 30 The valid loss 0.95257660108 15 / 30 The valid loss 0.935394146045 16 / 30 The valid loss 0.928010864183 17 / 30 The valid loss 0.941201621995 18 / 30 The valid loss 0.951099937161 19 / 30 The valid loss 0.951552718878 20 / 30 The valid loss 0.945320470631 21 / 30 The valid loss 0.931401072513 22 / 30 The valid loss 0.918460828337 23 / 30 The valid loss 0.906691018654 24 / 30 The valid loss 0.897944744676 25 / 30 The valid loss 0.889753683805 26 / 30 The valid loss 0.886297701643 27 / 30 The valid loss 0.875546260013 28 / 30 The valid loss 0.855954195772 29 / 30 The valid loss 0.833514231546 30 / 30 The valid loss 0.814645743867 Training 1 / 578 The train loss 0.585343718529 2 / 578 The train loss 0.462313354015 3 / 578 The train loss 0.40808703502 4 / 578 The train loss 0.396788150072 5 / 578 The train loss 0.389216268063 6 / 578 The train loss 0.355889151494 7 / 578 The train loss 0.331060524498 8 / 578 The train loss 0.303988270462 9 / 578 The train loss 0.282097337147 10 / 578 The train loss 0.265417277813 11 / 578 The train loss 0.251362566921 12 / 578 The train loss 0.241706396764 13 / 578 The train loss 0.237673442524 14 / 578 The train loss 0.243744613337 15 / 578 The train loss 0.245355279744 16 / 578 The train loss 0.245382917579 17 / 578 The train loss 0.243230500204 18 / 578 The train loss 0.237085392492 19 / 578 The train loss 0.229620682958 20 / 578 The train loss 0.2228332486 21 / 578 The train loss 0.215263814444 22 / 578 The train loss 0.208977617323 23 / 578 The train loss 0.204022323956 24 / 578 The train loss 0.198526477752 25 / 578 The train loss 0.194224468172 26 / 578 The train loss 0.190562610729 27 / 578 The train loss 0.187417198111 28 / 578 The train loss 0.186760930078 29 / 578 The train loss 0.18367674176 30 / 578 The train loss 0.181023115913 31 / 578 The train loss 0.178986553223 32 / 578 The train loss 0.177105327835 33 / 578 The train loss 0.176595344462 34 / 578 The train loss 0.177147342659 35 / 578 The train loss 0.179962765958 36 / 578 The train loss 0.183724402347 37 / 578 The train loss 0.186922682902 38 / 578 The train loss 0.190268734372 39 / 578 The train loss 0.191756355266 40 / 578 The train loss 0.192573825829 41 / 578 The train loss 0.191670372537 42 / 578 The train loss 0.191315604995 43 / 578 The train loss 0.192820837505 44 / 578 The train loss 0.194828347896 45 / 578 The train loss 0.195846639905 46 / 578 The train loss 0.195035662826 47 / 578 The train loss 0.194106152559 48 / 578 The train loss 0.192775078584 49 / 578 The train loss 0.191936972646 50 / 578 The train loss 0.193730748147 51 / 578 The train loss 0.195898819642 52 / 578 The train loss 0.197583833423 53 / 578 The train loss 0.198134990093 54 / 578 The train loss 0.198595368062 55 / 578 The train loss 0.198635620827 56 / 578 The train loss 0.198904960948 57 / 578 The train loss 0.200512867356 58 / 578 The train loss 0.20288128393 59 / 578 The train loss 0.20641789055 60 / 578 The train loss 0.212889377897 61 / 578 The train loss 0.223172348664 62 / 578 The train loss 0.232058742955 63 / 578 The train loss 0.240868764856 64 / 578 The train loss 0.247425542795 65 / 578 The train loss 0.252840477343 66 / 578 The train loss 0.255112083685 67 / 578 The train loss 0.256695397865 68 / 578 The train loss 0.257774278631 69 / 578 The train loss 0.258914866417 70 / 578 The train loss 0.262993855242 71 / 578 The train loss 0.268179963277 72 / 578 The train loss 0.274078358482 73 / 578 The train loss 0.278646455338 74 / 578 The train loss 0.282390812865 75 / 578 The train loss 0.286825447182 76 / 578 The train loss 0.289139685368 77 / 578 The train loss 0.290915244295 78 / 578 The train loss 0.292950536292 79 / 578 The train loss 0.2958471693 80 / 578 The train loss 0.297698792163 81 / 578 The train loss 0.298365494443 82 / 578 The train loss 0.298114898456 83 / 578 The train loss 0.298333116773 84 / 578 The train loss 0.298639304315 85 / 578 The train loss 0.298801211136 86 / 578 The train loss 0.298197670472 87 / 578 The train loss 0.297289403776 88 / 578 The train loss 0.296433810568 89 / 578 The train loss 0.298584444553 90 / 578 The train loss 0.299101209061 91 / 578 The train loss 0.297689134521 92 / 578 The train loss 0.295817790391 93 / 578 The train loss 0.294172048489 94 / 578 The train loss 0.292129690105 95 / 578 The train loss 0.29072479739 96 / 578 The train loss 0.28905224513 97 / 578 The train loss 0.287470736685 98 / 578 The train loss 0.286772334074 99 / 578 The train loss 0.286142425028 100 / 578 The train loss 0.285850884542 101 / 578 The train loss 0.286392595139 102 / 578 The train loss 0.285881089156 103 / 578 The train loss 0.285575984968 104 / 578 The train loss 0.285237567571 105 / 578 The train loss 0.28511041566 106 / 578 The train loss 0.28653531866 107 / 578 The train loss 0.286515146981 108 / 578 The train loss 0.286324569404 109 / 578 The train loss 0.286697586947 110 / 578 The train loss 0.286626179855 111 / 578 The train loss 0.285911899765 112 / 578 The train loss 0.285356475905 113 / 578 The train loss 0.285737374254 114 / 578 The train loss 0.286606711058 115 / 578 The train loss 0.288583622095 116 / 578 The train loss 0.290809825898 117 / 578 The train loss 0.292960387328 118 / 578 The train loss 0.296772112091 119 / 578 The train loss 0.30110399544 120 / 578 The train loss 0.305180313004 121 / 578 The train loss 0.309410048233 122 / 578 The train loss 0.315042939648 123 / 578 The train loss 0.322093726477 124 / 578 The train loss 0.332147059241 125 / 578 The train loss 0.341961291611 126 / 578 The train loss 0.352262098933 127 / 578 The train loss 0.362181171012 128 / 578 The train loss 0.373227197037 129 / 578 The train loss 0.385044057529 130 / 578 The train loss 0.397706343864 131 / 578 The train loss 0.406372062407 132 / 578 The train loss 0.410701497317 133 / 578 The train loss 0.412791541028 134 / 578 The train loss 0.414492094083 135 / 578 The train loss 0.418254302884 136 / 578 The train loss 0.422875770587 137 / 578 The train loss 0.42893164594 138 / 578 The train loss 0.43161276269 139 / 578 The train loss 0.433568137423 140 / 578 The train loss 0.434064354481 141 / 578 The train loss 0.434127709386 142 / 578 The train loss 0.434524265656 143 / 578 The train loss 0.436595447674 144 / 578 The train loss 0.438591304183 145 / 578 The train loss 0.440223726579 146 / 578 The train loss 0.441023014435 147 / 578 The train loss 0.441408781423 148 / 578 The train loss 0.44063657789 149 / 578 The train loss 0.439899768655 150 / 578 The train loss 0.439629734606 151 / 578 The train loss 0.437840452701 152 / 578 The train loss 0.436239349754 153 / 578 The train loss 0.434315428381 154 / 578 The train loss 0.432669918597 155 / 578 The train loss 0.432207704888 156 / 578 The train loss 0.431546575509 157 / 578 The train loss 0.431298322501 158 / 578 The train loss 0.431624040715 159 / 578 The train loss 0.431371950269 160 / 578 The train loss 0.430828844151 161 / 578 The train loss 0.43001143676 162 / 578 The train loss 0.429267270468 163 / 578 The train loss 0.428737255465 164 / 578 The train loss 0.42781678201 165 / 578 The train loss 0.42722208965 166 / 578 The train loss 0.426442416991 167 / 578 The train loss 0.425128001965 168 / 578 The train loss 0.427220278863 169 / 578 The train loss 0.435300116312 170 / 578 The train loss 0.443991631927 171 / 578 The train loss 0.452639916095 172 / 578 The train loss 0.460400343245 173 / 578 The train loss 0.467278242413 174 / 578 The train loss 0.474734508058 175 / 578 The train loss 0.482360176657 176 / 578 The train loss 0.489251132335 177 / 578 The train loss 0.493874296925 178 / 578 The train loss 0.496931500942 179 / 578 The train loss 0.501182838817 180 / 578 The train loss 0.504335781228 181 / 578 The train loss 0.506572643984 182 / 578 The train loss 0.507601623966 183 / 578 The train loss 0.508992800906 184 / 578 The train loss 0.513541966636 185 / 578 The train loss 0.518888305933 186 / 578 The train loss 0.5280717675 187 / 578 The train loss 0.54219678375 188 / 578 The train loss 0.556156072171 189 / 578 The train loss 0.56896173004 190 / 578 The train loss 0.585271643298 191 / 578 The train loss 0.598338391729 192 / 578 The train loss 0.60930641985 193 / 578 The train loss 0.618352442787 194 / 578 The train loss 0.625510782272 195 / 578 The train loss 0.629919822361 196 / 578 The train loss 0.6346041353 197 / 578 The train loss 0.642667642062 198 / 578 The train loss 0.647026412377 199 / 578 The train loss 0.651336540491 200 / 578 The train loss 0.656183385216 201 / 578 The train loss 0.665519410068 202 / 578 The train loss 0.679485105605 203 / 578 The train loss 0.691313555051 204 / 578 The train loss 0.700552313315 205 / 578 The train loss 0.70704857369 206 / 578 The train loss 0.710474272919 207 / 578 The train loss 0.713946059094 208 / 578 The train loss 0.72235004371 209 / 578 The train loss 0.733302968505 210 / 578 The train loss 0.741327935138 211 / 578 The train loss 0.747588806207 212 / 578 The train loss 0.755206133924 213 / 578 The train loss 0.761582147313 214 / 578 The train loss 0.766262442674 215 / 578 The train loss 0.770603861885 216 / 578 The train loss 0.773306164994 217 / 578 The train loss 0.775282746601 218 / 578 The train loss 0.777103054927 219 / 578 The train loss 0.779615687963 220 / 578 The train loss 0.782843977179 221 / 578 The train loss 0.787188258073 222 / 578 The train loss 0.791230087311 223 / 578 The train loss 0.795832266689 224 / 578 The train loss 0.800577453171 225 / 578 The train loss 0.807915422618 226 / 578 The train loss 0.815830796133 227 / 578 The train loss 0.823280258818 228 / 578 The train loss 0.829700264982 229 / 578 The train loss 0.836554363793 230 / 578 The train loss 0.842956098234 231 / 578 The train loss 0.849218531572 232 / 578 The train loss 0.854827444069 233 / 578 The train loss 0.860852092089 234 / 578 The train loss 0.867666707048 235 / 578 The train loss 0.876209175809 236 / 578 The train loss 0.881781518428 237 / 578 The train loss 0.883390287275 238 / 578 The train loss 0.88319379564 239 / 578 The train loss 0.883699575946 240 / 578 The train loss 0.884544350424 241 / 578 The train loss 0.886435453544 242 / 578 The train loss 0.889270208807 243 / 578 The train loss 0.892637893711 244 / 578 The train loss 0.896878663359 245 / 578 The train loss 0.902421862465 246 / 578 The train loss 0.907506617801 247 / 578 The train loss 0.909414389084 248 / 578 The train loss 0.911243158693 249 / 578 The train loss 0.91319980987 250 / 578 The train loss 0.917979822367 251 / 578 The train loss 0.922701588873 252 / 578 The train loss 0.927866860249 253 / 578 The train loss 0.93508884296 254 / 578 The train loss 0.944856646698 255 / 578 The train loss 0.95640000289 256 / 578 The train loss 0.96957715528 257 / 578 The train loss 0.981345127732 258 / 578 The train loss 0.990966529087 259 / 578 The train loss 0.997094047777 260 / 578 The train loss 1.00305731317 261 / 578 The train loss 1.00893379831 262 / 578 The train loss 1.01434189511 263 / 578 The train loss 1.01954731762 264 / 578 The train loss 1.02643555964 265 / 578 The train loss 1.03312621946 266 / 578 The train loss 1.03959643681 267 / 578 The train loss 1.04675892428 268 / 578 The train loss 1.05589739353 269 / 578 The train loss 1.06662363742 270 / 578 The train loss 1.07892972383 271 / 578 The train loss 1.08940908497 272 / 578 The train loss 1.09870944498 273 / 578 The train loss 1.10800024722 274 / 578 The train loss 1.11807166592 275 / 578 The train loss 1.12823872238 276 / 578 The train loss 1.13681612345 277 / 578 The train loss 1.14493474738 278 / 578 The train loss 1.15364747872 279 / 578 The train loss 1.16292020174 280 / 578 The train loss 1.17312531149 281 / 578 The train loss 1.18378082647 282 / 578 The train loss 1.19313383459 283 / 578 The train loss 1.20279518557 284 / 578 The train loss 1.21198764 285 / 578 The train loss 1.2206419177 286 / 578 The train loss 1.23000079287 287 / 578 The train loss 1.23857045026 288 / 578 The train loss 1.24618804721 289 / 578 The train loss 1.25366744799 290 / 578 The train loss 1.26210550458 291 / 578 The train loss 1.27029146134 292 / 578 The train loss 1.27889528113 293 / 578 The train loss 1.28721253556 294 / 578 The train loss 1.29544473847 295 / 578 The train loss 1.30342463576 296 / 578 The train loss 1.31085766104 297 / 578 The train loss 1.31716403192 298 / 578 The train loss 1.32337496202 299 / 578 The train loss 1.33037777905 300 / 578 The train loss 1.33794368284 301 / 578 The train loss 1.34524387238 302 / 578 The train loss 1.35337613925 303 / 578 The train loss 1.36025153624 304 / 578 The train loss 1.36692761451 305 / 578 The train loss 1.372619769 306 / 578 The train loss 1.37763075051 307 / 578 The train loss 1.38253845626 308 / 578 The train loss 1.387822622 309 / 578 The train loss 1.39162082674 310 / 578 The train loss 1.39346968414 311 / 578 The train loss 1.39336763526 312 / 578 The train loss 1.39233528133 313 / 578 The train loss 1.39012690529 314 / 578 The train loss 1.38849147563 315 / 578 The train loss 1.38756608355 316 / 578 The train loss 1.38733202556 317 / 578 The train loss 1.38742961076 318 / 578 The train loss 1.38712561241 319 / 578 The train loss 1.38626749909 320 / 578 The train loss 1.38450926428 321 / 578 The train loss 1.38173897595 322 / 578 The train loss 1.37843220831 323 / 578 The train loss 1.37493949388 324 / 578 The train loss 1.37120875438 325 / 578 The train loss 1.36740113311 326 / 578 The train loss 1.36340148009 327 / 578 The train loss 1.35950890327 328 / 578 The train loss 1.35567023444 329 / 578 The train loss 1.35193574657 330 / 578 The train loss 1.34820235527 331 / 578 The train loss 1.34443757487 332 / 578 The train loss 1.34076795749 333 / 578 The train loss 1.33751465022 334 / 578 The train loss 1.33487607499 335 / 578 The train loss 1.33340532873 336 / 578 The train loss 1.33284290193 337 / 578 The train loss 1.33362336528 338 / 578 The train loss 1.33441008852 339 / 578 The train loss 1.33384772093 340 / 578 The train loss 1.33220258395 341 / 578 The train loss 1.32920420812 342 / 578 The train loss 1.32624596728 343 / 578 The train loss 1.32366018006 344 / 578 The train loss 1.32134414113 345 / 578 The train loss 1.31934630193 346 / 578 The train loss 1.31738110098 347 / 578 The train loss 1.315658711 348 / 578 The train loss 1.31436841888 349 / 578 The train loss 1.31353573672 350 / 578 The train loss 1.31248559215 351 / 578 The train loss 1.31142269491 352 / 578 The train loss 1.31045504096 353 / 578 The train loss 1.30917800904 354 / 578 The train loss 1.30787609236 355 / 578 The train loss 1.30703879591 356 / 578 The train loss 1.30675557503 357 / 578 The train loss 1.30670126692 358 / 578 The train loss 1.30612039079 359 / 578 The train loss 1.30537047737 360 / 578 The train loss 1.30451588229 361 / 578 The train loss 1.30404378895 362 / 578 The train loss 1.3027147102 363 / 578 The train loss 1.30107430629 364 / 578 The train loss 1.29891518589 365 / 578 The train loss 1.29709193653 366 / 578 The train loss 1.29390060385 367 / 578 The train loss 1.29059374223 368 / 578 The train loss 1.28729642389 369 / 578 The train loss 1.28405926447 370 / 578 The train loss 1.28073724502 371 / 578 The train loss 1.27743563227 372 / 578 The train loss 1.2743522471 373 / 578 The train loss 1.27118582854 374 / 578 The train loss 1.26798363974 375 / 578 The train loss 1.26492847939 376 / 578 The train loss 1.26187980415 377 / 578 The train loss 1.25879136125 378 / 578 The train loss 1.25564540542 379 / 578 The train loss 1.25267879824 380 / 578 The train loss 1.24969333667 381 / 578 The train loss 1.24696620378 382 / 578 The train loss 1.24466157812 383 / 578 The train loss 1.24225888312 384 / 578 The train loss 1.23972309406 385 / 578 The train loss 1.23712460563 386 / 578 The train loss 1.23426490221 387 / 578 The train loss 1.23145795475 388 / 578 The train loss 1.22862453012 389 / 578 The train loss 1.22591708802 390 / 578 The train loss 1.22313721025 391 / 578 The train loss 1.22030233159 392 / 578 The train loss 1.2174824355 393 / 578 The train loss 1.21469589988 394 / 578 The train loss 1.21189152154 395 / 578 The train loss 1.20901375134 396 / 578 The train loss 1.20662361526 397 / 578 The train loss 1.20455908895 398 / 578 The train loss 1.20263308369 399 / 578 The train loss 1.20066003714 400 / 578 The train loss 1.1985044533 401 / 578 The train loss 1.1966008836 402 / 578 The train loss 1.19473777711 403 / 578 The train loss 1.19314864426 404 / 578 The train loss 1.19185901153 405 / 578 The train loss 1.19065263176 406 / 578 The train loss 1.18963039076 407 / 578 The train loss 1.18840136099 408 / 578 The train loss 1.18711806987 409 / 578 The train loss 1.18829193362 410 / 578 The train loss 1.19124084369 411 / 578 The train loss 1.19341780068 412 / 578 The train loss 1.19425649644 413 / 578 The train loss 1.19500137047 414 / 578 The train loss 1.19636872029 415 / 578 The train loss 1.19741165473 416 / 578 The train loss 1.19706923636 417 / 578 The train loss 1.19629992934 418 / 578 The train loss 1.19568010844 419 / 578 The train loss 1.19546228754 420 / 578 The train loss 1.1953869914 421 / 578 The train loss 1.19525362456 422 / 578 The train loss 1.19508150891 423 / 578 The train loss 1.19523271975 424 / 578 The train loss 1.19518704101 425 / 578 The train loss 1.19466058991 426 / 578 The train loss 1.19390793522 427 / 578 The train loss 1.19268459382 428 / 578 The train loss 1.19141954121 429 / 578 The train loss 1.19000382197 430 / 578 The train loss 1.18805064691 431 / 578 The train loss 1.18586670409 432 / 578 The train loss 1.18373236146 433 / 578 The train loss 1.18143530851 434 / 578 The train loss 1.17895752052 435 / 578 The train loss 1.17656120105 436 / 578 The train loss 1.17416485954 437 / 578 The train loss 1.17225742621 438 / 578 The train loss 1.17080561152 439 / 578 The train loss 1.16944397944 440 / 578 The train loss 1.16829274292 441 / 578 The train loss 1.16686050362 442 / 578 The train loss 1.16538976054 443 / 578 The train loss 1.16405047811 444 / 578 The train loss 1.16361880029 445 / 578 The train loss 1.16385622448 446 / 578 The train loss 1.16459056503 447 / 578 The train loss 1.16514413944 448 / 578 The train loss 1.16560732462 449 / 578 The train loss 1.16561721476 450 / 578 The train loss 1.16535328489 451 / 578 The train loss 1.16509171674 452 / 578 The train loss 1.16481526494 453 / 578 The train loss 1.16435777518 454 / 578 The train loss 1.16406211121 455 / 578 The train loss 1.16343261012 456 / 578 The train loss 1.16263761301 457 / 578 The train loss 1.16208680657 458 / 578 The train loss 1.16119730338 459 / 578 The train loss 1.160346963 460 / 578 The train loss 1.1597160896 461 / 578 The train loss 1.15908608049 462 / 578 The train loss 1.15824456179 463 / 578 The train loss 1.15693728373 464 / 578 The train loss 1.15491274582 465 / 578 The train loss 1.15281157345 466 / 578 The train loss 1.15066267651 467 / 578 The train loss 1.1484151253 468 / 578 The train loss 1.1463017559 469 / 578 The train loss 1.14435644202 470 / 578 The train loss 1.14249797563 471 / 578 The train loss 1.14059029584 472 / 578 The train loss 1.13864988472 473 / 578 The train loss 1.13670225808 474 / 578 The train loss 1.13465580534 475 / 578 The train loss 1.13274641274 476 / 578 The train loss 1.1307832833 477 / 578 The train loss 1.12879118 478 / 578 The train loss 1.12663684915 479 / 578 The train loss 1.12446244818 480 / 578 The train loss 1.1223501307 481 / 578 The train loss 1.12021102011 482 / 578 The train loss 1.11807282363 483 / 578 The train loss 1.11593379255 484 / 578 The train loss 1.11378283283 485 / 578 The train loss 1.11172900847 486 / 578 The train loss 1.10972423267 487 / 578 The train loss 1.1077076456 488 / 578 The train loss 1.10568853976 489 / 578 The train loss 1.10371117844 490 / 578 The train loss 1.10170628173 491 / 578 The train loss 1.09986660006 492 / 578 The train loss 1.09793974005 493 / 578 The train loss 1.09596523303 494 / 578 The train loss 1.09400381261 495 / 578 The train loss 1.09224624118 496 / 578 The train loss 1.09064922754 497 / 578 The train loss 1.08921878962 498 / 578 The train loss 1.0877183248 499 / 578 The train loss 1.08613630402 500 / 578 The train loss 1.08440936787 501 / 578 The train loss 1.08255109835 502 / 578 The train loss 1.0807157365 503 / 578 The train loss 1.07904081149 504 / 578 The train loss 1.07733898712 505 / 578 The train loss 1.07561216539 506 / 578 The train loss 1.07399495454 507 / 578 The train loss 1.07238437857 508 / 578 The train loss 1.07059418148 509 / 578 The train loss 1.06871563766 510 / 578 The train loss 1.06676501695 511 / 578 The train loss 1.06489622472 512 / 578 The train loss 1.06301970068 513 / 578 The train loss 1.06107279395 514 / 578 The train loss 1.05917411077 515 / 578 The train loss 1.05737564115 516 / 578 The train loss 1.05562693409 517 / 578 The train loss 1.05398672736 518 / 578 The train loss 1.05226717441 519 / 578 The train loss 1.05056269263 520 / 578 The train loss 1.04895661611 521 / 578 The train loss 1.04729033243 522 / 578 The train loss 1.04549012396 523 / 578 The train loss 1.04378293201 524 / 578 The train loss 1.04229309114 525 / 578 The train loss 1.04088503236 526 / 578 The train loss 1.03940143909 527 / 578 The train loss 1.03806128674 528 / 578 The train loss 1.03673409992 529 / 578 The train loss 1.03553018937 530 / 578 The train loss 1.03432523456 531 / 578 The train loss 1.03296697942 532 / 578 The train loss 1.03163403978 533 / 578 The train loss 1.03031578811 534 / 578 The train loss 1.02881214042 535 / 578 The train loss 1.02724316736 536 / 578 The train loss 1.02575736462 537 / 578 The train loss 1.02425262102 538 / 578 The train loss 1.0228261364 539 / 578 The train loss 1.02142637551 540 / 578 The train loss 1.02000555068 541 / 578 The train loss 1.01874587659 542 / 578 The train loss 1.01758744397 543 / 578 The train loss 1.01618784965 544 / 578 The train loss 1.01496716285 545 / 578 The train loss 1.01405166817 546 / 578 The train loss 1.01429527193 547 / 578 The train loss 1.01508697734 548 / 578 The train loss 1.01722104274 549 / 578 The train loss 1.0182741099 550 / 578 The train loss 1.01862213484 551 / 578 The train loss 1.01857212302 552 / 578 The train loss 1.01840906431 553 / 578 The train loss 1.01831253186 554 / 578 The train loss 1.01820187731 555 / 578 The train loss 1.0177571719 556 / 578 The train loss 1.01724343465 557 / 578 The train loss 1.01682720324 558 / 578 The train loss 1.01646947799 559 / 578 The train loss 1.01620951399 560 / 578 The train loss 1.01603848149 561 / 578 The train loss 1.01591600053 562 / 578 The train loss 1.01565385567 563 / 578 The train loss 1.01526856594 564 / 578 The train loss 1.0140200042 565 / 578 The train loss 1.01273205209 566 / 578 The train loss 1.01146081598 567 / 578 The train loss 1.0101418168 568 / 578 The train loss 1.00871619197 569 / 578 The train loss 1.00746192436 570 / 578 The train loss 1.00615323126 571 / 578 The train loss 1.00482371997 572 / 578 The train loss 1.00364650254 573 / 578 The train loss 1.00345989807 574 / 578 The train loss 1.00306046539 575 / 578 The train loss 1.0022145828 576 / 578 The train loss 1.00159908303 577 / 578 The train loss 1.00087527159 578 / 578 The train loss 1.00022822773 Starting epoch 5 Validation: 1 / 30 The valid loss 0.461230546236 2 / 30 The valid loss 0.436736941338 3 / 30 The valid loss 0.415480285883 4 / 30 The valid loss 0.394439183176 5 / 30 The valid loss 0.375673300028 6 / 30 The valid loss 0.416402960817 7 / 30 The valid loss 0.531373624291 8 / 30 The valid loss 0.676890257746 9 / 30 The valid loss 0.825435588757 10 / 30 The valid loss 0.90614079535 11 / 30 The valid loss 0.924591294744 12 / 30 The valid loss 0.936146430671 13 / 30 The valid loss 0.945419139587 14 / 30 The valid loss 0.950471641762 15 / 30 The valid loss 0.933322288593 16 / 30 The valid loss 0.925962751731 17 / 30 The valid loss 0.939150542021 18 / 30 The valid loss 0.949041212598 19 / 30 The valid loss 0.949490122105 20 / 30 The valid loss 0.943275134265 21 / 30 The valid loss 0.929360655092 22 / 30 The valid loss 0.916428743438 23 / 30 The valid loss 0.904666598724 24 / 30 The valid loss 0.895929029832 25 / 30 The valid loss 0.887731829882 26 / 30 The valid loss 0.884265987919 27 / 30 The valid loss 0.873517986801 28 / 30 The valid loss 0.853947629886 29 / 30 The valid loss 0.831562151683 30 / 30 The valid loss 0.812757950524 Training 1 / 578 The train loss 0.582132935524 2 / 578 The train loss 0.448431879282 3 / 578 The train loss 0.399144252141 4 / 578 The train loss 0.389005988836 5 / 578 The train loss 0.388296306133 6 / 578 The train loss 0.35618964086 7 / 578 The train loss 0.327719115785 8 / 578 The train loss 0.300493023358 9 / 578 The train loss 0.276602326996 10 / 578 The train loss 0.260494060069 11 / 578 The train loss 0.250355704942 12 / 578 The train loss 0.239170143381 13 / 578 The train loss 0.234192600044 14 / 578 The train loss 0.240105626306 15 / 578 The train loss 0.239837565521 16 / 578 The train loss 0.240404735785 17 / 578 The train loss 0.238166394041 18 / 578 The train loss 0.232135154721 19 / 578 The train loss 0.224954115325 20 / 578 The train loss 0.218580621853 21 / 578 The train loss 0.212597872885 22 / 578 The train loss 0.206349118528 23 / 578 The train loss 0.200017811973 24 / 578 The train loss 0.194751938339 25 / 578 The train loss 0.190755551904 26 / 578 The train loss 0.186484931323 27 / 578 The train loss 0.183370818281 28 / 578 The train loss 0.182498233792 29 / 578 The train loss 0.180666026499 30 / 578 The train loss 0.178627713645 31 / 578 The train loss 0.177055371505 32 / 578 The train loss 0.17541861895 33 / 578 The train loss 0.175474920846 34 / 578 The train loss 0.175616263357 35 / 578 The train loss 0.177261739011 36 / 578 The train loss 0.181307619541 37 / 578 The train loss 0.184524208207 38 / 578 The train loss 0.188644113411 39 / 578 The train loss 0.190066516304 40 / 578 The train loss 0.189698196109 41 / 578 The train loss 0.189173764844 42 / 578 The train loss 0.188662859567 43 / 578 The train loss 0.189515748603 44 / 578 The train loss 0.190528667701 45 / 578 The train loss 0.191897094167 46 / 578 The train loss 0.191186976093 47 / 578 The train loss 0.189931861818 48 / 578 The train loss 0.18873847827 49 / 578 The train loss 0.187548510198 50 / 578 The train loss 0.188160543367 51 / 578 The train loss 0.190352955212 52 / 578 The train loss 0.192585529545 53 / 578 The train loss 0.193338162424 54 / 578 The train loss 0.193497530495 55 / 578 The train loss 0.193573845178 56 / 578 The train loss 0.193111288694 57 / 578 The train loss 0.194523569196 58 / 578 The train loss 0.197926333688 59 / 578 The train loss 0.202945572782 60 / 578 The train loss 0.210021043755 61 / 578 The train loss 0.220150317264 62 / 578 The train loss 0.2291065252 63 / 578 The train loss 0.237412439393 64 / 578 The train loss 0.243615467974 65 / 578 The train loss 0.247310299999 66 / 578 The train loss 0.248960224008 67 / 578 The train loss 0.25054300493 68 / 578 The train loss 0.251366400927 69 / 578 The train loss 0.251738080458 70 / 578 The train loss 0.254298541322 71 / 578 The train loss 0.259018193616 72 / 578 The train loss 0.264551444962 73 / 578 The train loss 0.269194560457 74 / 578 The train loss 0.273202489306 75 / 578 The train loss 0.277395443072 76 / 578 The train loss 0.28001037941 77 / 578 The train loss 0.281681462106 78 / 578 The train loss 0.283592917168 79 / 578 The train loss 0.286177480098 80 / 578 The train loss 0.288520865468 81 / 578 The train loss 0.289858016289 82 / 578 The train loss 0.289810401592 83 / 578 The train loss 0.289161226791 84 / 578 The train loss 0.289335800024 85 / 578 The train loss 0.289418127212 86 / 578 The train loss 0.289128422694 87 / 578 The train loss 0.289002205406 88 / 578 The train loss 0.289408787365 89 / 578 The train loss 0.291862684647 90 / 578 The train loss 0.293371378341 91 / 578 The train loss 0.291803094835 92 / 578 The train loss 0.290424425276 93 / 578 The train loss 0.288656540415 94 / 578 The train loss 0.286664895436 95 / 578 The train loss 0.28447191029 96 / 578 The train loss 0.283162211417 97 / 578 The train loss 0.281875540032 98 / 578 The train loss 0.281830633819 99 / 578 The train loss 0.281638785783 100 / 578 The train loss 0.282743790187 101 / 578 The train loss 0.283059781178 102 / 578 The train loss 0.282459646208 103 / 578 The train loss 0.281386018125 104 / 578 The train loss 0.282093625552 105 / 578 The train loss 0.281536257586 106 / 578 The train loss 0.282151551214 107 / 578 The train loss 0.282299039555 108 / 578 The train loss 0.281969943291 109 / 578 The train loss 0.282134350868 110 / 578 The train loss 0.281327969276 111 / 578 The train loss 0.28246908994 112 / 578 The train loss 0.282428036743 113 / 578 The train loss 0.283101863203 114 / 578 The train loss 0.284844584796 115 / 578 The train loss 0.287257703653 116 / 578 The train loss 0.28938136086 117 / 578 The train loss 0.291825228426 118 / 578 The train loss 0.295531690279 119 / 578 The train loss 0.299722706727 120 / 578 The train loss 0.303502569254 121 / 578 The train loss 0.307430242365 122 / 578 The train loss 0.31208047286 123 / 578 The train loss 0.319736217732 124 / 578 The train loss 0.329529415846 125 / 578 The train loss 0.33943394056 126 / 578 The train loss 0.349327053963 127 / 578 The train loss 0.35954285856 128 / 578 The train loss 0.37056486556 129 / 578 The train loss 0.382232438369 130 / 578 The train loss 0.394676464022 131 / 578 The train loss 0.404021704157 132 / 578 The train loss 0.409212871624 133 / 578 The train loss 0.411783546125 134 / 578 The train loss 0.413983563573 135 / 578 The train loss 0.417718202676 136 / 578 The train loss 0.422250373097 137 / 578 The train loss 0.427444964131 138 / 578 The train loss 0.430065015223 139 / 578 The train loss 0.432265419031 140 / 578 The train loss 0.433026832901 141 / 578 The train loss 0.432689964269 142 / 578 The train loss 0.433048189027 143 / 578 The train loss 0.43495894721 144 / 578 The train loss 0.436902853272 145 / 578 The train loss 0.438647340007 146 / 578 The train loss 0.439136222949 147 / 578 The train loss 0.439235384494 148 / 578 The train loss 0.43819644355 149 / 578 The train loss 0.437092828396 150 / 578 The train loss 0.43666754894 151 / 578 The train loss 0.434582355722 152 / 578 The train loss 0.432685307105 153 / 578 The train loss 0.430708398154 154 / 578 The train loss 0.429205465747 155 / 578 The train loss 0.428321703139 156 / 578 The train loss 0.427328385914 157 / 578 The train loss 0.427013894863 158 / 578 The train loss 0.427428427683 159 / 578 The train loss 0.427086458946 160 / 578 The train loss 0.426325508137 161 / 578 The train loss 0.425169632232 162 / 578 The train loss 0.423830819466 163 / 578 The train loss 0.423245787186 164 / 578 The train loss 0.422340370647 165 / 578 The train loss 0.420755057014 166 / 578 The train loss 0.41943774581 167 / 578 The train loss 0.418414811732 168 / 578 The train loss 0.422630970522 169 / 578 The train loss 0.432023467948 170 / 578 The train loss 0.440525356308 171 / 578 The train loss 0.44788290728 172 / 578 The train loss 0.453494780827 173 / 578 The train loss 0.460560606933 174 / 578 The train loss 0.467441598424 175 / 578 The train loss 0.474373846586 176 / 578 The train loss 0.481559067139 177 / 578 The train loss 0.485517402679 178 / 578 The train loss 0.488814993374 179 / 578 The train loss 0.493416012853 180 / 578 The train loss 0.497026740035 181 / 578 The train loss 0.499509761532 182 / 578 The train loss 0.500793411366 183 / 578 The train loss 0.502440396086 184 / 578 The train loss 0.507261747434 185 / 578 The train loss 0.51282304769 186 / 578 The train loss 0.522293023705 187 / 578 The train loss 0.536089002947 188 / 578 The train loss 0.548989909305 189 / 578 The train loss 0.56033882343 190 / 578 The train loss 0.575617464769 191 / 578 The train loss 0.58805124253 192 / 578 The train loss 0.599481362403 193 / 578 The train loss 0.609604065456 194 / 578 The train loss 0.617128235336 195 / 578 The train loss 0.621699742954 196 / 578 The train loss 0.626745128681 197 / 578 The train loss 0.634932737835 198 / 578 The train loss 0.638970404732 199 / 578 The train loss 0.643697575877 200 / 578 The train loss 0.648983875681 201 / 578 The train loss 0.658584328007 202 / 578 The train loss 0.67250636089 203 / 578 The train loss 0.685277359991 204 / 578 The train loss 0.69446918677 205 / 578 The train loss 0.701522816845 206 / 578 The train loss 0.704877590254 207 / 578 The train loss 0.708145301861 208 / 578 The train loss 0.71560990368 209 / 578 The train loss 0.726729917915 210 / 578 The train loss 0.734966610887 211 / 578 The train loss 0.741311984198 212 / 578 The train loss 0.748387823848 213 / 578 The train loss 0.754964898047 214 / 578 The train loss 0.75978339124 215 / 578 The train loss 0.763985357884 216 / 578 The train loss 0.767010177882 217 / 578 The train loss 0.76894223266 218 / 578 The train loss 0.770651343919 219 / 578 The train loss 0.773193442811 220 / 578 The train loss 0.776395847191 221 / 578 The train loss 0.780862102293 222 / 578 The train loss 0.78505275125 223 / 578 The train loss 0.78983030206 224 / 578 The train loss 0.794907123 225 / 578 The train loss 0.802034885767 226 / 578 The train loss 0.809563288795 227 / 578 The train loss 0.816855103056 228 / 578 The train loss 0.823133803398 229 / 578 The train loss 0.829679573614 230 / 578 The train loss 0.836130328946 231 / 578 The train loss 0.842979885428 232 / 578 The train loss 0.849099662942 233 / 578 The train loss 0.85537587127 234 / 578 The train loss 0.862081866122 235 / 578 The train loss 0.869979309671 236 / 578 The train loss 0.875780707841 237 / 578 The train loss 0.876974163318 238 / 578 The train loss 0.876910394371 239 / 578 The train loss 0.87721460124 240 / 578 The train loss 0.877873441965 241 / 578 The train loss 0.880153239407 242 / 578 The train loss 0.882853555177 243 / 578 The train loss 0.885990793215 244 / 578 The train loss 0.889688344168 245 / 578 The train loss 0.894631502473 246 / 578 The train loss 0.899599296185 247 / 578 The train loss 0.901615134995 248 / 578 The train loss 0.902999828314 249 / 578 The train loss 0.904463979988 250 / 578 The train loss 0.90846284999 251 / 578 The train loss 0.912889284338 252 / 578 The train loss 0.917504822821 253 / 578 The train loss 0.923586684621 254 / 578 The train loss 0.932019064174 255 / 578 The train loss 0.94288483544 256 / 578 The train loss 0.954596680953 257 / 578 The train loss 0.965953659353 258 / 578 The train loss 0.974955101707 259 / 578 The train loss 0.979863125745 260 / 578 The train loss 0.984746284109 261 / 578 The train loss 0.989621294313 262 / 578 The train loss 0.993342265728 263 / 578 The train loss 0.99786090025 264 / 578 The train loss 1.00391233861 265 / 578 The train loss 1.00890652179 266 / 578 The train loss 1.01490610514 267 / 578 The train loss 1.02097403234 268 / 578 The train loss 1.02922911154 269 / 578 The train loss 1.03761806266 270 / 578 The train loss 1.04790735853 271 / 578 The train loss 1.05608421049 272 / 578 The train loss 1.0635739003 273 / 578 The train loss 1.07258208234 274 / 578 The train loss 1.08079021159 275 / 578 The train loss 1.08750010047 276 / 578 The train loss 1.09449603583 277 / 578 The train loss 1.09941025236 278 / 578 The train loss 1.10544685142 279 / 578 The train loss 1.11144496966 280 / 578 The train loss 1.119591024 281 / 578 The train loss 1.12715879529 282 / 578 The train loss 1.1340138125 283 / 578 The train loss 1.14055399077 284 / 578 The train loss 1.14629233378 285 / 578 The train loss 1.15126987498 286 / 578 The train loss 1.15636934841 287 / 578 The train loss 1.16203751123 288 / 578 The train loss 1.16375794211 289 / 578 The train loss 1.16811647084 290 / 578 The train loss 1.17476661484 291 / 578 The train loss 1.18084425893 292 / 578 The train loss 1.187385045 293 / 578 The train loss 1.19384720577 294 / 578 The train loss 1.19911402455 295 / 578 The train loss 1.20641432083 296 / 578 The train loss 1.21038349689 297 / 578 The train loss 1.2137238565 298 / 578 The train loss 1.21612831961 299 / 578 The train loss 1.22244613297 300 / 578 The train loss 1.22865731469 301 / 578 The train loss 1.2314282041 302 / 578 The train loss 1.2344944871 303 / 578 The train loss 1.23860490825 304 / 578 The train loss 1.24189584576 305 / 578 The train loss 1.24318238 306 / 578 The train loss 1.24494926438 307 / 578 The train loss 1.24811829875 308 / 578 The train loss 1.25066973706 309 / 578 The train loss 1.25226667814 310 / 578 The train loss 1.25382169515 311 / 578 The train loss 1.25336696396 312 / 578 The train loss 1.25354699362 313 / 578 The train loss 1.2531006503 314 / 578 The train loss 1.25285373477 315 / 578 The train loss 1.25140485793 316 / 578 The train loss 1.25201446602 317 / 578 The train loss 1.25147678152 318 / 578 The train loss 1.25080567703 319 / 578 The train loss 1.24930462747 320 / 578 The train loss 1.24775640565 321 / 578 The train loss 1.24580023121 322 / 578 The train loss 1.24292586328 323 / 578 The train loss 1.24055277657 324 / 578 The train loss 1.23783521282 325 / 578 The train loss 1.23528209027 326 / 578 The train loss 1.2321357995 327 / 578 The train loss 1.22906685665 328 / 578 The train loss 1.22608702227 329 / 578 The train loss 1.22358439502 330 / 578 The train loss 1.22049611763 331 / 578 The train loss 1.21771184538 332 / 578 The train loss 1.21477235796 333 / 578 The train loss 1.21198366727 334 / 578 The train loss 1.20969062014 335 / 578 The train loss 1.20869395952 336 / 578 The train loss 1.20870476293 337 / 578 The train loss 1.20901575752 338 / 578 The train loss 1.2084077696 339 / 578 The train loss 1.20673595546 340 / 578 The train loss 1.20451949761 341 / 578 The train loss 1.20222146163 342 / 578 The train loss 1.19953774244 343 / 578 The train loss 1.197387169 344 / 578 The train loss 1.19536920174 345 / 578 The train loss 1.19345562377 346 / 578 The train loss 1.19144878667 347 / 578 The train loss 1.19020442537 348 / 578 The train loss 1.18951940961 349 / 578 The train loss 1.18912393945 350 / 578 The train loss 1.18866390446 351 / 578 The train loss 1.18806608566 352 / 578 The train loss 1.18755140982 353 / 578 The train loss 1.18675169219 354 / 578 The train loss 1.18576518304 355 / 578 The train loss 1.18540458626 356 / 578 The train loss 1.18545615586 357 / 578 The train loss 1.18520304115 358 / 578 The train loss 1.18481829093 359 / 578 The train loss 1.18407912089 360 / 578 The train loss 1.18353027715 361 / 578 The train loss 1.18297745445 362 / 578 The train loss 1.18191196145 363 / 578 The train loss 1.18059249732 364 / 578 The train loss 1.17893037 365 / 578 The train loss 1.17724369467 366 / 578 The train loss 1.17428417831 367 / 578 The train loss 1.1712766987 368 / 578 The train loss 1.16830275023 369 / 578 The train loss 1.16531745297 370 / 578 The train loss 1.16244394847 371 / 578 The train loss 1.15950130488 372 / 578 The train loss 1.15656645154 373 / 578 The train loss 1.15370800125 374 / 578 The train loss 1.15074882388 375 / 578 The train loss 1.14792893965 376 / 578 The train loss 1.1450617385 377 / 578 The train loss 1.14220512588 378 / 578 The train loss 1.13934254801 379 / 578 The train loss 1.13663499621 380 / 578 The train loss 1.13392936665 381 / 578 The train loss 1.13137629369 382 / 578 The train loss 1.12910984838 383 / 578 The train loss 1.12687131302 384 / 578 The train loss 1.12443930267 385 / 578 The train loss 1.12195210077 386 / 578 The train loss 1.11929320508 387 / 578 The train loss 1.11670331436 388 / 578 The train loss 1.11419656956 389 / 578 The train loss 1.11174984815 390 / 578 The train loss 1.10923835562 391 / 578 The train loss 1.10663435529 392 / 578 The train loss 1.10407097371 393 / 578 The train loss 1.10154352937 394 / 578 The train loss 1.09893849574 395 / 578 The train loss 1.09631820407 396 / 578 The train loss 1.09405910091 397 / 578 The train loss 1.09204885757 398 / 578 The train loss 1.09018500145 399 / 578 The train loss 1.08834701167 400 / 578 The train loss 1.08651458043 401 / 578 The train loss 1.08471399742 402 / 578 The train loss 1.0831017903 403 / 578 The train loss 1.08162904779 404 / 578 The train loss 1.08028722919 405 / 578 The train loss 1.07916365938 406 / 578 The train loss 1.0779782876 407 / 578 The train loss 1.07676772143 408 / 578 The train loss 1.07537290106 409 / 578 The train loss 1.07639127258 410 / 578 The train loss 1.07920584681 411 / 578 The train loss 1.08167799143 412 / 578 The train loss 1.08240599629 413 / 578 The train loss 1.08285089904 414 / 578 The train loss 1.08418799745 415 / 578 The train loss 1.08519994468 416 / 578 The train loss 1.08495744464 417 / 578 The train loss 1.08429952279 418 / 578 The train loss 1.08383945688 419 / 578 The train loss 1.08363551871 420 / 578 The train loss 1.08370440335 421 / 578 The train loss 1.08353117316 422 / 578 The train loss 1.08340948159 423 / 578 The train loss 1.08342780362 424 / 578 The train loss 1.08345440047 425 / 578 The train loss 1.08309531158 426 / 578 The train loss 1.08242588706 427 / 578 The train loss 1.08134772853 428 / 578 The train loss 1.08004534372 429 / 578 The train loss 1.07852387917 430 / 578 The train loss 1.07666384303 431 / 578 The train loss 1.07465132859 432 / 578 The train loss 1.07271542479 433 / 578 The train loss 1.0705208274 434 / 578 The train loss 1.0682158273 435 / 578 The train loss 1.06594396236 436 / 578 The train loss 1.06366810605 437 / 578 The train loss 1.06171831062 438 / 578 The train loss 1.06023736015 439 / 578 The train loss 1.05895260135 440 / 578 The train loss 1.05754527242 441 / 578 The train loss 1.05605243416 442 / 578 The train loss 1.05452791513 443 / 578 The train loss 1.05327022798 444 / 578 The train loss 1.05282436133 445 / 578 The train loss 1.05274643733 446 / 578 The train loss 1.05339519562 447 / 578 The train loss 1.05389272256 448 / 578 The train loss 1.05408902755 449 / 578 The train loss 1.05399791696 450 / 578 The train loss 1.05364797098 451 / 578 The train loss 1.05331491389 452 / 578 The train loss 1.05288622438 453 / 578 The train loss 1.05256277477 454 / 578 The train loss 1.0522379603 455 / 578 The train loss 1.05173345905 456 / 578 The train loss 1.05103784978 457 / 578 The train loss 1.0503618499 458 / 578 The train loss 1.049491989 459 / 578 The train loss 1.04873845023 460 / 578 The train loss 1.04827719029 461 / 578 The train loss 1.04756830903 462 / 578 The train loss 1.04665108027 463 / 578 The train loss 1.04547644005 464 / 578 The train loss 1.04363276182 465 / 578 The train loss 1.0417930286 466 / 578 The train loss 1.03984053413 467 / 578 The train loss 1.03793260199 468 / 578 The train loss 1.03605065207 469 / 578 The train loss 1.03428802109 470 / 578 The train loss 1.03267478511 471 / 578 The train loss 1.03115570829 472 / 578 The train loss 1.02964330066 473 / 578 The train loss 1.02812673834 474 / 578 The train loss 1.02643466184 475 / 578 The train loss 1.02483001746 476 / 578 The train loss 1.02322915803 477 / 578 The train loss 1.02163807444 478 / 578 The train loss 1.01979048687 479 / 578 The train loss 1.01791541598 480 / 578 The train loss 1.01599302187 481 / 578 The train loss 1.01411876192 482 / 578 The train loss 1.01227686645 483 / 578 The train loss 1.01033906056 484 / 578 The train loss 1.00850689818 485 / 578 The train loss 1.00674899485 486 / 578 The train loss 1.00497971175 487 / 578 The train loss 1.00322534484 488 / 578 The train loss 1.00149334006 489 / 578 The train loss 0.999660111469 490 / 578 The train loss 0.997848596264 491 / 578 The train loss 0.996131700686 492 / 578 The train loss 0.99444733473 493 / 578 The train loss 0.992776160648 494 / 578 The train loss 0.991111688685 495 / 578 The train loss 0.989499506302 496 / 578 The train loss 0.988208363832 497 / 578 The train loss 0.987076818501 498 / 578 The train loss 0.985867385801 499 / 578 The train loss 0.984570942603 500 / 578 The train loss 0.983163507141 501 / 578 The train loss 0.981575428941 502 / 578 The train loss 0.98002152297 503 / 578 The train loss 0.978608573444 504 / 578 The train loss 0.977236917132 505 / 578 The train loss 0.97585499851 506 / 578 The train loss 0.974422150015 507 / 578 The train loss 0.973040729151 508 / 578 The train loss 0.971459148298 509 / 578 The train loss 0.969831796176 510 / 578 The train loss 0.968122155913 511 / 578 The train loss 0.966493793612 512 / 578 The train loss 0.964803461167 513 / 578 The train loss 0.963071356282 514 / 578 The train loss 0.961446031964 515 / 578 The train loss 0.959841334364 516 / 578 The train loss 0.958336808719 517 / 578 The train loss 0.956902591162 518 / 578 The train loss 0.955396422296 519 / 578 The train loss 0.953856428902 520 / 578 The train loss 0.952443784196 521 / 578 The train loss 0.951016005944 522 / 578 The train loss 0.949354096598 523 / 578 The train loss 0.947858928359 524 / 578 The train loss 0.946563517696 525 / 578 The train loss 0.945419923046 526 / 578 The train loss 0.944221482793 527 / 578 The train loss 0.943043761033 528 / 578 The train loss 0.941959189399 529 / 578 The train loss 0.94092326988 530 / 578 The train loss 0.939856985809 531 / 578 The train loss 0.9387009028 532 / 578 The train loss 0.937528599341 533 / 578 The train loss 0.936446215636 534 / 578 The train loss 0.935097118644 535 / 578 The train loss 0.933657364039 536 / 578 The train loss 0.932278288063 537 / 578 The train loss 0.930931127112 538 / 578 The train loss 0.929693775778 539 / 578 The train loss 0.928385246531 540 / 578 The train loss 0.927127370383 541 / 578 The train loss 0.926030053098 542 / 578 The train loss 0.924888370036 543 / 578 The train loss 0.923674679572 544 / 578 The train loss 0.922592499163 545 / 578 The train loss 0.921863782003 546 / 578 The train loss 0.922291823721 547 / 578 The train loss 0.923032868204 548 / 578 The train loss 0.92536913384 549 / 578 The train loss 0.926495817766 550 / 578 The train loss 0.926963922469 551 / 578 The train loss 0.927139250781 552 / 578 The train loss 0.927122660123 553 / 578 The train loss 0.92721796729 554 / 578 The train loss 0.92725023211 555 / 578 The train loss 0.927007694214 556 / 578 The train loss 0.926790242177 557 / 578 The train loss 0.926559074703 558 / 578 The train loss 0.926387544314 559 / 578 The train loss 0.926310535014 560 / 578 The train loss 0.92629376898 561 / 578 The train loss 0.926219143153 562 / 578 The train loss 0.92617868585 563 / 578 The train loss 0.925966021573 564 / 578 The train loss 0.92480639155 565 / 578 The train loss 0.923666107213 566 / 578 The train loss 0.922544666905 567 / 578 The train loss 0.921370283249 568 / 578 The train loss 0.920167003148 569 / 578 The train loss 0.918965144721 570 / 578 The train loss 0.917763255256 571 / 578 The train loss 0.916571087033 572 / 578 The train loss 0.915604348763 573 / 578 The train loss 0.915645065713 574 / 578 The train loss 0.915371119645 575 / 578 The train loss 0.914657374756 576 / 578 The train loss 0.914263237455 577 / 578 The train loss 0.91368963481 578 / 578 The train loss 0.91314507725 Starting epoch 6 Validation: 1 / 30 The valid loss 0.459952056408 2 / 30 The valid loss 0.434771224856 3 / 30 The valid loss 0.413475165764 4 / 30 The valid loss 0.392620511353 5 / 30 The valid loss 0.374136942625 6 / 30 The valid loss 0.414793416858 7 / 30 The valid loss 0.529141379254 8 / 30 The valid loss 0.674032632262 9 / 30 The valid loss 0.821537842353 10 / 30 The valid loss 0.901765134931 11 / 30 The valid loss 0.920721208507 12 / 30 The valid loss 0.932454191148 13 / 30 The valid loss 0.941873791126 14 / 30 The valid loss 0.947054132819 15 / 30 The valid loss 0.930054499706 16 / 30 The valid loss 0.922820614651 17 / 30 The valid loss 0.936101964291 18 / 30 The valid loss 0.946072871486 19 / 30 The valid loss 0.946592388969 20 / 30 The valid loss 0.940458412468 21 / 30 The valid loss 0.926593634344 22 / 30 The valid loss 0.913664613258 23 / 30 The valid loss 0.901916233094 24 / 30 The valid loss 0.893182296306 25 / 30 The valid loss 0.885033475161 26 / 30 The valid loss 0.881569006122 27 / 30 The valid loss 0.870916983596 28 / 30 The valid loss 0.851460305708 29 / 30 The valid loss 0.829148044874 30 / 30 The valid loss 0.81040789187 Training 1 / 578 The train loss 0.626258134842 2 / 578 The train loss 0.469520762563 3 / 578 The train loss 0.417056510846 4 / 578 The train loss 0.400502204895 5 / 578 The train loss 0.391342496872 6 / 578 The train loss 0.364060387015 7 / 578 The train loss 0.335100099444 8 / 578 The train loss 0.305032305419 9 / 578 The train loss 0.280792375406 10 / 578 The train loss 0.26172843948 11 / 578 The train loss 0.249848959798 12 / 578 The train loss 0.23897072052 13 / 578 The train loss 0.235132181874 14 / 578 The train loss 0.241454255368 15 / 578 The train loss 0.241284985344 16 / 578 The train loss 0.24050160218 17 / 578 The train loss 0.237975292346 18 / 578 The train loss 0.232441717552 19 / 578 The train loss 0.225546866655 20 / 578 The train loss 0.217566535249 21 / 578 The train loss 0.211292358736 22 / 578 The train loss 0.205454060638 23 / 578 The train loss 0.19977289503 24 / 578 The train loss 0.194803595232 25 / 578 The train loss 0.189755877554 26 / 578 The train loss 0.185417285332 27 / 578 The train loss 0.182329990522 28 / 578 The train loss 0.180298913004 29 / 578 The train loss 0.178197857378 30 / 578 The train loss 0.176303203901 31 / 578 The train loss 0.174394806306 32 / 578 The train loss 0.172116106143 33 / 578 The train loss 0.171303769178 34 / 578 The train loss 0.171873682781 35 / 578 The train loss 0.174616888804 36 / 578 The train loss 0.179937825642 37 / 578 The train loss 0.183072878098 38 / 578 The train loss 0.186232080781 39 / 578 The train loss 0.188586738056 40 / 578 The train loss 0.189050966688 41 / 578 The train loss 0.188130311668 42 / 578 The train loss 0.188054444357 43 / 578 The train loss 0.189129201305 44 / 578 The train loss 0.190273269164 45 / 578 The train loss 0.191504500144 46 / 578 The train loss 0.190703652313 47 / 578 The train loss 0.189939449284 48 / 578 The train loss 0.1892057308 49 / 578 The train loss 0.188260815429 50 / 578 The train loss 0.189859092087 51 / 578 The train loss 0.191855368515 52 / 578 The train loss 0.194433761187 53 / 578 The train loss 0.194730689644 54 / 578 The train loss 0.194298796218 55 / 578 The train loss 0.193438162722 56 / 578 The train loss 0.193081024103 57 / 578 The train loss 0.193620159046 58 / 578 The train loss 0.196102046889 59 / 578 The train loss 0.200393029315 60 / 578 The train loss 0.208172313496 61 / 578 The train loss 0.218138281561 62 / 578 The train loss 0.227183920842 63 / 578 The train loss 0.235106911096 64 / 578 The train loss 0.240876827505 65 / 578 The train loss 0.244041193105 66 / 578 The train loss 0.246015141069 67 / 578 The train loss 0.2468233292 68 / 578 The train loss 0.247659648714 69 / 578 The train loss 0.248882072235 70 / 578 The train loss 0.251957239956 71 / 578 The train loss 0.257630234558 72 / 578 The train loss 0.262513785416 73 / 578 The train loss 0.266398730037 74 / 578 The train loss 0.270061275141 75 / 578 The train loss 0.273765782217 76 / 578 The train loss 0.275541133492 77 / 578 The train loss 0.276649526671 78 / 578 The train loss 0.278269960808 79 / 578 The train loss 0.280348683554 80 / 578 The train loss 0.281295906845 81 / 578 The train loss 0.281815891005 82 / 578 The train loss 0.281508429112 83 / 578 The train loss 0.280780241881 84 / 578 The train loss 0.280491194289 85 / 578 The train loss 0.279900973597 86 / 578 The train loss 0.27903565236 87 / 578 The train loss 0.278234420066 88 / 578 The train loss 0.277801375582 89 / 578 The train loss 0.279567214294 90 / 578 The train loss 0.279884049379 91 / 578 The train loss 0.278079998035 92 / 578 The train loss 0.276357650595 93 / 578 The train loss 0.274837410578 94 / 578 The train loss 0.273846844092 95 / 578 The train loss 0.27220859559 96 / 578 The train loss 0.270765156796 97 / 578 The train loss 0.269129866024 98 / 578 The train loss 0.26861349843 99 / 578 The train loss 0.268609931072 100 / 578 The train loss 0.26743417576 101 / 578 The train loss 0.267487018563 102 / 578 The train loss 0.267717247354 103 / 578 The train loss 0.267777486768 104 / 578 The train loss 0.267229186371 105 / 578 The train loss 0.266487713939 106 / 578 The train loss 0.266726985574 107 / 578 The train loss 0.266430011698 108 / 578 The train loss 0.266756931389 109 / 578 The train loss 0.267737312054 110 / 578 The train loss 0.266730102626 111 / 578 The train loss 0.266335605098 112 / 578 The train loss 0.266939818194 113 / 578 The train loss 0.268134913481 114 / 578 The train loss 0.268964647855 115 / 578 The train loss 0.269793399132 116 / 578 The train loss 0.271318275995 117 / 578 The train loss 0.274420666516 118 / 578 The train loss 0.277484621663 119 / 578 The train loss 0.281675567772 120 / 578 The train loss 0.286401254063 121 / 578 The train loss 0.290937251419 122 / 578 The train loss 0.295066399286 123 / 578 The train loss 0.301269595822 124 / 578 The train loss 0.311118993668 125 / 578 The train loss 0.321526767373 126 / 578 The train loss 0.330510113566 127 / 578 The train loss 0.339749211988 128 / 578 The train loss 0.350221931585 129 / 578 The train loss 0.361592473563 130 / 578 The train loss 0.373508133911 131 / 578 The train loss 0.383364157481 132 / 578 The train loss 0.388415358848 133 / 578 The train loss 0.390655621214 134 / 578 The train loss 0.392294957678 135 / 578 The train loss 0.395341619849 136 / 578 The train loss 0.399704518673 137 / 578 The train loss 0.404818758586 138 / 578 The train loss 0.40770730603 139 / 578 The train loss 0.409587182265 140 / 578 The train loss 0.409963371392 141 / 578 The train loss 0.409952374321 142 / 578 The train loss 0.410406488033 143 / 578 The train loss 0.412653651375 144 / 578 The train loss 0.414519245529 145 / 578 The train loss 0.415755140884 146 / 578 The train loss 0.416668889559 147 / 578 The train loss 0.416685256524 148 / 578 The train loss 0.415793278632 149 / 578 The train loss 0.415002511152 150 / 578 The train loss 0.414653503199 151 / 578 The train loss 0.412760277852 152 / 578 The train loss 0.41109188499 153 / 578 The train loss 0.409352030045 154 / 578 The train loss 0.407863284473 155 / 578 The train loss 0.407207848564 156 / 578 The train loss 0.40641749784 157 / 578 The train loss 0.405932205308 158 / 578 The train loss 0.405717311969 159 / 578 The train loss 0.405378154051 160 / 578 The train loss 0.405140279047 161 / 578 The train loss 0.404039347764 162 / 578 The train loss 0.40280511267 163 / 578 The train loss 0.402231153809 164 / 578 The train loss 0.401276499578 165 / 578 The train loss 0.400046837691 166 / 578 The train loss 0.399284768535 167 / 578 The train loss 0.398322839312 168 / 578 The train loss 0.402088097252 169 / 578 The train loss 0.409080207083 170 / 578 The train loss 0.418147975645 171 / 578 The train loss 0.426038029685 172 / 578 The train loss 0.432021375399 173 / 578 The train loss 0.439495753088 174 / 578 The train loss 0.446543577844 175 / 578 The train loss 0.453743089352 176 / 578 The train loss 0.460645410232 177 / 578 The train loss 0.464722543197 178 / 578 The train loss 0.467978336922 179 / 578 The train loss 0.471368353543 180 / 578 The train loss 0.474435893777 181 / 578 The train loss 0.476035121808 182 / 578 The train loss 0.477150754369 183 / 578 The train loss 0.478471988293 184 / 578 The train loss 0.482424078185 185 / 578 The train loss 0.487273587166 186 / 578 The train loss 0.496840115276 187 / 578 The train loss 0.510053575756 188 / 578 The train loss 0.522179329411 189 / 578 The train loss 0.5332194167 190 / 578 The train loss 0.548266572466 191 / 578 The train loss 0.560599230737 192 / 578 The train loss 0.570863040552 193 / 578 The train loss 0.579550814521 194 / 578 The train loss 0.585925786535 195 / 578 The train loss 0.590357146278 196 / 578 The train loss 0.59521789309 197 / 578 The train loss 0.603144983123 198 / 578 The train loss 0.608079020197 199 / 578 The train loss 0.613664462114 200 / 578 The train loss 0.619491140023 201 / 578 The train loss 0.629357362046 202 / 578 The train loss 0.643611476342 203 / 578 The train loss 0.656529606637 204 / 578 The train loss 0.666514713755 205 / 578 The train loss 0.673831521374 206 / 578 The train loss 0.677141942229 207 / 578 The train loss 0.680632187908 208 / 578 The train loss 0.688985941287 209 / 578 The train loss 0.699917076664 210 / 578 The train loss 0.707556602856 211 / 578 The train loss 0.714978116172 212 / 578 The train loss 0.722497121132 213 / 578 The train loss 0.729479351864 214 / 578 The train loss 0.734869399709 215 / 578 The train loss 0.73911262813 216 / 578 The train loss 0.742445988029 217 / 578 The train loss 0.744476401105 218 / 578 The train loss 0.745947302823 219 / 578 The train loss 0.748742615047 220 / 578 The train loss 0.752305953408 221 / 578 The train loss 0.756283698915 222 / 578 The train loss 0.760048026236 223 / 578 The train loss 0.764167091037 224 / 578 The train loss 0.768853510397 225 / 578 The train loss 0.775882431997 226 / 578 The train loss 0.783653104529 227 / 578 The train loss 0.791183712527 228 / 578 The train loss 0.797559074469 229 / 578 The train loss 0.804443729301 230 / 578 The train loss 0.810645569213 231 / 578 The train loss 0.817466986966 232 / 578 The train loss 0.823251690651 233 / 578 The train loss 0.82949677794 234 / 578 The train loss 0.835849558505 235 / 578 The train loss 0.844566896431 236 / 578 The train loss 0.850402769712 237 / 578 The train loss 0.85213080671 238 / 578 The train loss 0.852138659095 239 / 578 The train loss 0.852494107442 240 / 578 The train loss 0.853393411947 241 / 578 The train loss 0.855683421507 242 / 578 The train loss 0.858967236066 243 / 578 The train loss 0.862039527591 244 / 578 The train loss 0.866234537336 245 / 578 The train loss 0.871899873687 246 / 578 The train loss 0.877069572426 247 / 578 The train loss 0.879010814706 248 / 578 The train loss 0.880025933166 249 / 578 The train loss 0.881962255601 250 / 578 The train loss 0.885544009507 251 / 578 The train loss 0.889997037105 252 / 578 The train loss 0.894745759843 253 / 578 The train loss 0.901529010871 254 / 578 The train loss 0.911021716306 255 / 578 The train loss 0.922118551883 256 / 578 The train loss 0.934409862675 257 / 578 The train loss 0.94431962699 258 / 578 The train loss 0.953123939419 259 / 578 The train loss 0.957908379353 260 / 578 The train loss 0.9630701719 261 / 578 The train loss 0.96770321849 262 / 578 The train loss 0.971195155257 263 / 578 The train loss 0.975661049154 264 / 578 The train loss 0.980106482734 265 / 578 The train loss 0.985355922242 266 / 578 The train loss 0.989567464288 267 / 578 The train loss 0.994119546666 268 / 578 The train loss 1.00214390062 269 / 578 The train loss 1.01025119308 270 / 578 The train loss 1.01890777642 271 / 578 The train loss 1.02653645045 272 / 578 The train loss 1.03272844309 273 / 578 The train loss 1.0381694179 274 / 578 The train loss 1.04191952275 275 / 578 The train loss 1.04795377542 276 / 578 The train loss 1.0539328558 277 / 578 The train loss 1.05984201656 278 / 578 The train loss 1.06413089613 279 / 578 The train loss 1.06799290839 280 / 578 The train loss 1.07521876166 281 / 578 The train loss 1.08078816591 282 / 578 The train loss 1.08444678768 283 / 578 The train loss 1.08891721818 284 / 578 The train loss 1.09300113044 285 / 578 The train loss 1.09678135488 286 / 578 The train loss 1.1014388523 287 / 578 The train loss 1.1060225771 288 / 578 The train loss 1.11135709436 289 / 578 The train loss 1.11371115525 290 / 578 The train loss 1.11782922277 291 / 578 The train loss 1.12179887034 292 / 578 The train loss 1.12580084336 293 / 578 The train loss 1.12722115453 294 / 578 The train loss 1.13140665415 295 / 578 The train loss 1.13411708833 296 / 578 The train loss 1.13848575935 297 / 578 The train loss 1.14188662508 298 / 578 The train loss 1.14357030857 299 / 578 The train loss 1.14569755686 300 / 578 The train loss 1.14933136448 301 / 578 The train loss 1.15072991825 302 / 578 The train loss 1.15419754351 303 / 578 The train loss 1.15667274629 304 / 578 The train loss 1.15885392091 305 / 578 The train loss 1.16005375371 306 / 578 The train loss 1.16211431697 307 / 578 The train loss 1.16368582723 308 / 578 The train loss 1.16543680957 309 / 578 The train loss 1.16651215508 310 / 578 The train loss 1.16714355816 311 / 578 The train loss 1.16818321312 312 / 578 The train loss 1.16873756419 313 / 578 The train loss 1.16890989539 314 / 578 The train loss 1.17072394039 315 / 578 The train loss 1.17117193374 316 / 578 The train loss 1.17120111993 317 / 578 The train loss 1.17039444982 318 / 578 The train loss 1.1712668949 319 / 578 The train loss 1.17115553032 320 / 578 The train loss 1.17075355747 321 / 578 The train loss 1.16939661831 322 / 578 The train loss 1.16728804779 323 / 578 The train loss 1.16520236922 324 / 578 The train loss 1.16291201625 325 / 578 The train loss 1.16073999052 326 / 578 The train loss 1.15837373191 327 / 578 The train loss 1.15598972023 328 / 578 The train loss 1.15321450258 329 / 578 The train loss 1.1506859407 330 / 578 The train loss 1.14784266443 331 / 578 The train loss 1.14499884686 332 / 578 The train loss 1.14216040586 333 / 578 The train loss 1.13945774058 334 / 578 The train loss 1.13738098027 335 / 578 The train loss 1.13669693363 336 / 578 The train loss 1.13684158382 337 / 578 The train loss 1.13767265318 338 / 578 The train loss 1.13717579524 339 / 578 The train loss 1.13562874826 340 / 578 The train loss 1.1331803356 341 / 578 The train loss 1.13098871769 342 / 578 The train loss 1.12846548193 343 / 578 The train loss 1.1264920816 344 / 578 The train loss 1.12454630461 345 / 578 The train loss 1.12279991231 346 / 578 The train loss 1.1211200552 347 / 578 The train loss 1.11975783702 348 / 578 The train loss 1.11895000412 349 / 578 The train loss 1.11866756932 350 / 578 The train loss 1.11835445157 351 / 578 The train loss 1.11779547052 352 / 578 The train loss 1.11747490115 353 / 578 The train loss 1.11694029683 354 / 578 The train loss 1.1162335136 355 / 578 The train loss 1.11609059228 356 / 578 The train loss 1.11612202888 357 / 578 The train loss 1.1162654022 358 / 578 The train loss 1.11601660946 359 / 578 The train loss 1.1155539287 360 / 578 The train loss 1.114942361 361 / 578 The train loss 1.11425416663 362 / 578 The train loss 1.11349526468 363 / 578 The train loss 1.11230593747 364 / 578 The train loss 1.11087613635 365 / 578 The train loss 1.10960286292 366 / 578 The train loss 1.10681066654 367 / 578 The train loss 1.10398172916 368 / 578 The train loss 1.10119483442 369 / 578 The train loss 1.09849651479 370 / 578 The train loss 1.09569512177 371 / 578 The train loss 1.09293303307 372 / 578 The train loss 1.09022313741 373 / 578 The train loss 1.08746018343 374 / 578 The train loss 1.08474196149 375 / 578 The train loss 1.08205149194 376 / 578 The train loss 1.07935818921 377 / 578 The train loss 1.07668537818 378 / 578 The train loss 1.07396881966 379 / 578 The train loss 1.07132465911 380 / 578 The train loss 1.06876405513 381 / 578 The train loss 1.06647755127 382 / 578 The train loss 1.06433188709 383 / 578 The train loss 1.06214759287 384 / 578 The train loss 1.05988134417 385 / 578 The train loss 1.05747652201 386 / 578 The train loss 1.05494947168 387 / 578 The train loss 1.05246034168 388 / 578 The train loss 1.05006798484 389 / 578 The train loss 1.0478431096 390 / 578 The train loss 1.04543387059 391 / 578 The train loss 1.04296153107 392 / 578 The train loss 1.04054705266 393 / 578 The train loss 1.03814188563 394 / 578 The train loss 1.03569907033 395 / 578 The train loss 1.03327586547 396 / 578 The train loss 1.03106585151 397 / 578 The train loss 1.02927340817 398 / 578 The train loss 1.02760832234 399 / 578 The train loss 1.02590458089 400 / 578 The train loss 1.0241279914 401 / 578 The train loss 1.02248311494 402 / 578 The train loss 1.02103905934 403 / 578 The train loss 1.01987504632 404 / 578 The train loss 1.01880574505 405 / 578 The train loss 1.01764666561 406 / 578 The train loss 1.0164965024 407 / 578 The train loss 1.01530629175 408 / 578 The train loss 1.01404401001 409 / 578 The train loss 1.01506874131 410 / 578 The train loss 1.01807729077 411 / 578 The train loss 1.0200835311 412 / 578 The train loss 1.02064235536 413 / 578 The train loss 1.02116851554 414 / 578 The train loss 1.02257534802 415 / 578 The train loss 1.02377907467 416 / 578 The train loss 1.02363798737 417 / 578 The train loss 1.02307484631 418 / 578 The train loss 1.02251503312 419 / 578 The train loss 1.02222650829 420 / 578 The train loss 1.0220807301 421 / 578 The train loss 1.02180820872 422 / 578 The train loss 1.02164363492 423 / 578 The train loss 1.02182694077 424 / 578 The train loss 1.02173966955 425 / 578 The train loss 1.02139416875 426 / 578 The train loss 1.02070902181 427 / 578 The train loss 1.0197775876 428 / 578 The train loss 1.01850411782 429 / 578 The train loss 1.01706780947 430 / 578 The train loss 1.01521692472 431 / 578 The train loss 1.01334207187 432 / 578 The train loss 1.01141691748 433 / 578 The train loss 1.00929763087 434 / 578 The train loss 1.00712946792 435 / 578 The train loss 1.00495595852 436 / 578 The train loss 1.00275682491 437 / 578 The train loss 1.00083997158 438 / 578 The train loss 0.999337946588 439 / 578 The train loss 0.998074857828 440 / 578 The train loss 0.996712285831 441 / 578 The train loss 0.995176069058 442 / 578 The train loss 0.993612558063 443 / 578 The train loss 0.99246000861 444 / 578 The train loss 0.991908858161 445 / 578 The train loss 0.991682629848 446 / 578 The train loss 0.992062653603 447 / 578 The train loss 0.992307328213 448 / 578 The train loss 0.992363576586 449 / 578 The train loss 0.992228973491 450 / 578 The train loss 0.991805375359 451 / 578 The train loss 0.991360899557 452 / 578 The train loss 0.99098056971 453 / 578 The train loss 0.990569032523 454 / 578 The train loss 0.990098626414 455 / 578 The train loss 0.989425018667 456 / 578 The train loss 0.98871640803 457 / 578 The train loss 0.988056181109 458 / 578 The train loss 0.987181325089 459 / 578 The train loss 0.986209080183 460 / 578 The train loss 0.985429023344 461 / 578 The train loss 0.984693140187 462 / 578 The train loss 0.983678822595 463 / 578 The train loss 0.982777446903 464 / 578 The train loss 0.981207830069 465 / 578 The train loss 0.979650685488 466 / 578 The train loss 0.977975995284 467 / 578 The train loss 0.976287518606 468 / 578 The train loss 0.974684534204 469 / 578 The train loss 0.973286710862 470 / 578 The train loss 0.97199663955 471 / 578 The train loss 0.970721052497 472 / 578 The train loss 0.969397984737 473 / 578 The train loss 0.968062688736 474 / 578 The train loss 0.966676954769 475 / 578 The train loss 0.965331079685 476 / 578 The train loss 0.964069623758 477 / 578 The train loss 0.962710489369 478 / 578 The train loss 0.961094240771 479 / 578 The train loss 0.959276399783 480 / 578 The train loss 0.957576576659 481 / 578 The train loss 0.955858945328 482 / 578 The train loss 0.954143200649 483 / 578 The train loss 0.952366387043 484 / 578 The train loss 0.950708390491 485 / 578 The train loss 0.949097340109 486 / 578 The train loss 0.94749135813 487 / 578 The train loss 0.945863470076 488 / 578 The train loss 0.944236460562 489 / 578 The train loss 0.942557233068 490 / 578 The train loss 0.940843172425 491 / 578 The train loss 0.939259775142 492 / 578 The train loss 0.937661493326 493 / 578 The train loss 0.93611747074 494 / 578 The train loss 0.934498631663 495 / 578 The train loss 0.933029995625 496 / 578 The train loss 0.931885480603 497 / 578 The train loss 0.930800301051 498 / 578 The train loss 0.929698712937 499 / 578 The train loss 0.928585990181 500 / 578 The train loss 0.927231447779 501 / 578 The train loss 0.925735812932 502 / 578 The train loss 0.924335389966 503 / 578 The train loss 0.923057803984 504 / 578 The train loss 0.921881728163 505 / 578 The train loss 0.920627906587 506 / 578 The train loss 0.91940762149 507 / 578 The train loss 0.91815698005 508 / 578 The train loss 0.916668948395 509 / 578 The train loss 0.915110206169 510 / 578 The train loss 0.913503255732 511 / 578 The train loss 0.911928485481 512 / 578 The train loss 0.910346303826 513 / 578 The train loss 0.908743317444 514 / 578 The train loss 0.907194568149 515 / 578 The train loss 0.905732043804 516 / 578 The train loss 0.904343139243 517 / 578 The train loss 0.903025216252 518 / 578 The train loss 0.901651127926 519 / 578 The train loss 0.900228165255 520 / 578 The train loss 0.898828255965 521 / 578 The train loss 0.897469913555 522 / 578 The train loss 0.895987816362 523 / 578 The train loss 0.894623073812 524 / 578 The train loss 0.893512066887 525 / 578 The train loss 0.892351353857 526 / 578 The train loss 0.8912464186 527 / 578 The train loss 0.890271830046 528 / 578 The train loss 0.889245353308 529 / 578 The train loss 0.888250182511 530 / 578 The train loss 0.88730746334 531 / 578 The train loss 0.886309666324 532 / 578 The train loss 0.885212654073 533 / 578 The train loss 0.884115399615 534 / 578 The train loss 0.882910789127 535 / 578 The train loss 0.881545245459 536 / 578 The train loss 0.880239088095 537 / 578 The train loss 0.878993700787 538 / 578 The train loss 0.877946935573 539 / 578 The train loss 0.876822710486 540 / 578 The train loss 0.875625971463 541 / 578 The train loss 0.874557314162 542 / 578 The train loss 0.87354319755 543 / 578 The train loss 0.872392494273 544 / 578 The train loss 0.871415895581 545 / 578 The train loss 0.870853705807 546 / 578 The train loss 0.871336918181 547 / 578 The train loss 0.872200541573 548 / 578 The train loss 0.874533966299 549 / 578 The train loss 0.875729706125 550 / 578 The train loss 0.876247303709 551 / 578 The train loss 0.876587738582 552 / 578 The train loss 0.87659375705 553 / 578 The train loss 0.876809659321 554 / 578 The train loss 0.876976281196 555 / 578 The train loss 0.876792581024 556 / 578 The train loss 0.876568013501 557 / 578 The train loss 0.876495969794 558 / 578 The train loss 0.876282651478 559 / 578 The train loss 0.876299742942 560 / 578 The train loss 0.876280471157 561 / 578 The train loss 0.876363332159 562 / 578 The train loss 0.876373158344 563 / 578 The train loss 0.876207868018 564 / 578 The train loss 0.875168889186 565 / 578 The train loss 0.874067795705 566 / 578 The train loss 0.873055170109 567 / 578 The train loss 0.871960157394 568 / 578 The train loss 0.870877961989 569 / 578 The train loss 0.869802049631 570 / 578 The train loss 0.868651676498 571 / 578 The train loss 0.867598094013 572 / 578 The train loss 0.866742386899 573 / 578 The train loss 0.866877861366 574 / 578 The train loss 0.866814655093 575 / 578 The train loss 0.866220947355 576 / 578 The train loss 0.865813545682 577 / 578 The train loss 0.865420120641 578 / 578 The train loss 0.865068930971 Starting epoch 7 Validation: 1 / 30 The valid loss 0.460619390011 2 / 30 The valid loss 0.434220254421 3 / 30 The valid loss 0.412182827791 4 / 30 The valid loss 0.389269173145 5 / 30 The valid loss 0.369528317451 6 / 30 The valid loss 0.409326553345 7 / 30 The valid loss 0.523039494242 8 / 30 The valid loss 0.667545869946 9 / 30 The valid loss 0.815114484893 10 / 30 The valid loss 0.895310270786 11 / 30 The valid loss 0.913519014012 12 / 30 The valid loss 0.924832443396 13 / 30 The valid loss 0.933901511706 14 / 30 The valid loss 0.938810084547 15 / 30 The valid loss 0.921830789248 16 / 30 The valid loss 0.914595603943 17 / 30 The valid loss 0.927754864973 18 / 30 The valid loss 0.937581181526 19 / 30 The valid loss 0.937997190576 20 / 30 The valid loss 0.931871595979 21 / 30 The valid loss 0.917975854306 22 / 30 The valid loss 0.905017736283 23 / 30 The valid loss 0.893271974895 24 / 30 The valid loss 0.884557388723 25 / 30 The valid loss 0.876311607361 26 / 30 The valid loss 0.872761923533 27 / 30 The valid loss 0.862045413918 28 / 30 The valid loss 0.842631978648 29 / 30 The valid loss 0.820557589161 30 / 30 The valid loss 0.802116503318 Training 1 / 578 The train loss 0.545511305332 2 / 578 The train loss 0.429003953934 3 / 578 The train loss 0.370063503583 4 / 578 The train loss 0.349467344582 5 / 578 The train loss 0.339624977112 6 / 578 The train loss 0.314291191598 7 / 578 The train loss 0.290816645537 8 / 578 The train loss 0.268902912736 9 / 578 The train loss 0.249025240541 10 / 578 The train loss 0.234442157298 11 / 578 The train loss 0.223806204444 12 / 578 The train loss 0.21508252124 13 / 578 The train loss 0.212106788388 14 / 578 The train loss 0.221281546567 15 / 578 The train loss 0.223086054126 16 / 578 The train loss 0.225264702924 17 / 578 The train loss 0.223527194823 18 / 578 The train loss 0.218922045496 19 / 578 The train loss 0.212653293029 20 / 578 The train loss 0.206082719564 21 / 578 The train loss 0.199617974815 22 / 578 The train loss 0.193872708489 23 / 578 The train loss 0.187811118429 24 / 578 The train loss 0.183247463778 25 / 578 The train loss 0.1799192518 26 / 578 The train loss 0.17634577734 27 / 578 The train loss 0.173805183283 28 / 578 The train loss 0.172112909279 29 / 578 The train loss 0.169889001497 30 / 578 The train loss 0.167571807653 31 / 578 The train loss 0.167233176049 32 / 578 The train loss 0.165692131035 33 / 578 The train loss 0.165454849149 34 / 578 The train loss 0.165961085873 35 / 578 The train loss 0.169078024796 36 / 578 The train loss 0.173634384241 37 / 578 The train loss 0.177712702268 38 / 578 The train loss 0.181239784548 39 / 578 The train loss 0.182243063282 40 / 578 The train loss 0.182575846091 41 / 578 The train loss 0.182306158107 42 / 578 The train loss 0.181886247226 43 / 578 The train loss 0.181905237048 44 / 578 The train loss 0.183511325582 45 / 578 The train loss 0.184663411644 46 / 578 The train loss 0.184230680375 47 / 578 The train loss 0.182690494555 48 / 578 The train loss 0.181242321618 49 / 578 The train loss 0.180258127804 50 / 578 The train loss 0.18105233103 51 / 578 The train loss 0.183164953309 52 / 578 The train loss 0.185183171756 53 / 578 The train loss 0.186071251642 54 / 578 The train loss 0.185196018053 55 / 578 The train loss 0.184975505417 56 / 578 The train loss 0.184185754242 57 / 578 The train loss 0.185501535995 58 / 578 The train loss 0.187575330765 59 / 578 The train loss 0.190976651291 60 / 578 The train loss 0.198099903514 61 / 578 The train loss 0.206946667345 62 / 578 The train loss 0.215683314348 63 / 578 The train loss 0.223295758877 64 / 578 The train loss 0.228637626627 65 / 578 The train loss 0.231928134194 66 / 578 The train loss 0.233423070248 67 / 578 The train loss 0.234719742368 68 / 578 The train loss 0.236055682468 69 / 578 The train loss 0.236717572031 70 / 578 The train loss 0.240349719567 71 / 578 The train loss 0.244396343407 72 / 578 The train loss 0.24973008554 73 / 578 The train loss 0.254251834669 74 / 578 The train loss 0.257896160959 75 / 578 The train loss 0.261919086973 76 / 578 The train loss 0.264263897154 77 / 578 The train loss 0.265038719618 78 / 578 The train loss 0.267000014774 79 / 578 The train loss 0.27007003204 80 / 578 The train loss 0.271396568976 81 / 578 The train loss 0.271790295287 82 / 578 The train loss 0.271369457063 83 / 578 The train loss 0.272375095142 84 / 578 The train loss 0.273438992777 85 / 578 The train loss 0.274170907631 86 / 578 The train loss 0.274185239402 87 / 578 The train loss 0.274348327312 88 / 578 The train loss 0.273539550941 89 / 578 The train loss 0.275812881046 90 / 578 The train loss 0.276074615121 91 / 578 The train loss 0.274979636892 92 / 578 The train loss 0.273244884594 93 / 578 The train loss 0.272232928824 94 / 578 The train loss 0.27042172112 95 / 578 The train loss 0.268941737398 96 / 578 The train loss 0.267093756779 97 / 578 The train loss 0.265903165276 98 / 578 The train loss 0.2650885899 99 / 578 The train loss 0.265391801719 100 / 578 The train loss 0.265731997266 101 / 578 The train loss 0.26610298772 102 / 578 The train loss 0.266246990903 103 / 578 The train loss 0.265873862051 104 / 578 The train loss 0.26489859669 105 / 578 The train loss 0.264466143435 106 / 578 The train loss 0.264142333407 107 / 578 The train loss 0.264409099853 108 / 578 The train loss 0.266263992215 109 / 578 The train loss 0.266422220395 110 / 578 The train loss 0.266779589179 111 / 578 The train loss 0.267304745947 112 / 578 The train loss 0.268521816669 113 / 578 The train loss 0.269409461161 114 / 578 The train loss 0.270660116811 115 / 578 The train loss 0.272250848814 116 / 578 The train loss 0.273850000431 117 / 578 The train loss 0.276869644276 118 / 578 The train loss 0.280957738537 119 / 578 The train loss 0.285262548435 120 / 578 The train loss 0.289083508961 121 / 578 The train loss 0.293286650449 122 / 578 The train loss 0.298588149188 123 / 578 The train loss 0.306035353219 124 / 578 The train loss 0.315486314136 125 / 578 The train loss 0.325150739968 126 / 578 The train loss 0.334152727196 127 / 578 The train loss 0.343433174855 128 / 578 The train loss 0.354024140455 129 / 578 The train loss 0.364989220923 130 / 578 The train loss 0.377591337894 131 / 578 The train loss 0.386137346559 132 / 578 The train loss 0.390411526247 133 / 578 The train loss 0.39241076709 134 / 578 The train loss 0.394466338596 135 / 578 The train loss 0.397818302501 136 / 578 The train loss 0.402303895803 137 / 578 The train loss 0.407256952439 138 / 578 The train loss 0.410087786129 139 / 578 The train loss 0.412111470007 140 / 578 The train loss 0.412757311389 141 / 578 The train loss 0.412641345662 142 / 578 The train loss 0.41289737616 143 / 578 The train loss 0.414768344925 144 / 578 The train loss 0.416472123367 145 / 578 The train loss 0.417820146628 146 / 578 The train loss 0.418131531969 147 / 578 The train loss 0.418334907466 148 / 578 The train loss 0.417272276824 149 / 578 The train loss 0.416108605896 150 / 578 The train loss 0.4154146038 151 / 578 The train loss 0.413404357256 152 / 578 The train loss 0.4115686298 153 / 578 The train loss 0.409518391917 154 / 578 The train loss 0.407770374717 155 / 578 The train loss 0.407056624995 156 / 578 The train loss 0.406273902752 157 / 578 The train loss 0.405801393187 158 / 578 The train loss 0.405626184531 159 / 578 The train loss 0.405011961248 160 / 578 The train loss 0.403722262243 161 / 578 The train loss 0.402568935098 162 / 578 The train loss 0.401260563575 163 / 578 The train loss 0.400270597289 164 / 578 The train loss 0.399045935082 165 / 578 The train loss 0.397833828357 166 / 578 The train loss 0.397198410323 167 / 578 The train loss 0.39588331797 168 / 578 The train loss 0.398427947424 169 / 578 The train loss 0.406791889235 170 / 578 The train loss 0.414628817535 171 / 578 The train loss 0.422990673946 172 / 578 The train loss 0.430196692561 173 / 578 The train loss 0.437270524659 174 / 578 The train loss 0.444335230937 175 / 578 The train loss 0.451915257105 176 / 578 The train loss 0.458308998622 177 / 578 The train loss 0.463793858086 178 / 578 The train loss 0.466858294363 179 / 578 The train loss 0.470929344065 180 / 578 The train loss 0.473800587447 181 / 578 The train loss 0.475969945981 182 / 578 The train loss 0.477021281659 183 / 578 The train loss 0.478057501868 184 / 578 The train loss 0.48243410444 185 / 578 The train loss 0.487505414156 186 / 578 The train loss 0.496259207487 187 / 578 The train loss 0.509137237893 188 / 578 The train loss 0.520426184692 189 / 578 The train loss 0.53151787916 190 / 578 The train loss 0.546055668396 191 / 578 The train loss 0.557923008751 192 / 578 The train loss 0.568048843609 193 / 578 The train loss 0.576563370776 194 / 578 The train loss 0.582923927988 195 / 578 The train loss 0.587099945354 196 / 578 The train loss 0.591702700992 197 / 578 The train loss 0.599638228603 198 / 578 The train loss 0.60383791104 199 / 578 The train loss 0.608599296879 200 / 578 The train loss 0.613748274557 201 / 578 The train loss 0.623322906297 202 / 578 The train loss 0.637567614085 203 / 578 The train loss 0.649499539194 204 / 578 The train loss 0.658569645173 205 / 578 The train loss 0.665691927693 206 / 578 The train loss 0.668504213932 207 / 578 The train loss 0.671415218517 208 / 578 The train loss 0.679020000299 209 / 578 The train loss 0.689931137247 210 / 578 The train loss 0.697878515259 211 / 578 The train loss 0.704395419719 212 / 578 The train loss 0.711268966033 213 / 578 The train loss 0.717052097157 214 / 578 The train loss 0.721241473142 215 / 578 The train loss 0.725325554575 216 / 578 The train loss 0.728429297158 217 / 578 The train loss 0.729713607597 218 / 578 The train loss 0.731261651553 219 / 578 The train loss 0.734127955996 220 / 578 The train loss 0.73709320978 221 / 578 The train loss 0.741094921071 222 / 578 The train loss 0.744932890576 223 / 578 The train loss 0.749300098306 224 / 578 The train loss 0.753774350343 225 / 578 The train loss 0.760051363938 226 / 578 The train loss 0.767130707316 227 / 578 The train loss 0.77427364981 228 / 578 The train loss 0.779749283763 229 / 578 The train loss 0.785972107023 230 / 578 The train loss 0.792471732361 231 / 578 The train loss 0.798758053773 232 / 578 The train loss 0.804546664597 233 / 578 The train loss 0.810237965455 234 / 578 The train loss 0.817244743785 235 / 578 The train loss 0.825973847667 236 / 578 The train loss 0.831475991408 237 / 578 The train loss 0.832988646054 238 / 578 The train loss 0.832970769559 239 / 578 The train loss 0.832883550626 240 / 578 The train loss 0.833125152284 241 / 578 The train loss 0.833982081427 242 / 578 The train loss 0.835903193925 243 / 578 The train loss 0.837994110529 244 / 578 The train loss 0.841014914917 245 / 578 The train loss 0.844513819747 246 / 578 The train loss 0.84764455595 247 / 578 The train loss 0.848014033366 248 / 578 The train loss 0.848030672288 249 / 578 The train loss 0.847505554946 250 / 578 The train loss 0.849056098789 251 / 578 The train loss 0.85021436912 252 / 578 The train loss 0.851853143926 253 / 578 The train loss 0.854175425654 254 / 578 The train loss 0.859700600316 255 / 578 The train loss 0.863873027207 256 / 578 The train loss 0.871159572009 257 / 578 The train loss 0.875674376547 258 / 578 The train loss 0.879970574419 259 / 578 The train loss 0.88010503297 260 / 578 The train loss 0.881572027476 261 / 578 The train loss 0.881715787182 262 / 578 The train loss 0.881129397776 263 / 578 The train loss 0.88137170443 264 / 578 The train loss 0.882024892888 265 / 578 The train loss 0.882359072932 266 / 578 The train loss 0.882301543076 267 / 578 The train loss 0.882625712075 268 / 578 The train loss 0.885666331332 269 / 578 The train loss 0.888889642093 270 / 578 The train loss 0.893513029234 271 / 578 The train loss 0.897383062245 272 / 578 The train loss 0.900520601337 273 / 578 The train loss 0.901237227186 274 / 578 The train loss 0.901157161914 275 / 578 The train loss 0.901826804199 276 / 578 The train loss 0.90272652787 277 / 578 The train loss 0.902915344207 278 / 578 The train loss 0.903003902181 279 / 578 The train loss 0.902093905782 280 / 578 The train loss 0.901792515894 281 / 578 The train loss 0.900871580685 282 / 578 The train loss 0.900280198175 283 / 578 The train loss 0.900246971356 284 / 578 The train loss 0.900014238537 285 / 578 The train loss 0.899649936353 286 / 578 The train loss 0.901402937444 287 / 578 The train loss 0.902008688486 288 / 578 The train loss 0.902120693147 289 / 578 The train loss 0.902586755327 290 / 578 The train loss 0.904663736938 291 / 578 The train loss 0.90559707011 292 / 578 The train loss 0.906538382657 293 / 578 The train loss 0.906885468385 294 / 578 The train loss 0.907847096193 295 / 578 The train loss 0.909635413981 296 / 578 The train loss 0.910379938454 297 / 578 The train loss 0.911050771097 298 / 578 The train loss 0.911489290954 299 / 578 The train loss 0.912472469626 300 / 578 The train loss 0.912978546893 301 / 578 The train loss 0.913839856528 302 / 578 The train loss 0.916290571629 303 / 578 The train loss 0.917880774602 304 / 578 The train loss 0.918287293961 305 / 578 The train loss 0.918796144817 306 / 578 The train loss 0.919292569818 307 / 578 The train loss 0.919177208229 308 / 578 The train loss 0.919467212197 309 / 578 The train loss 0.919410697923 310 / 578 The train loss 0.919654840111 311 / 578 The train loss 0.920311311122 312 / 578 The train loss 0.921116824023 313 / 578 The train loss 0.922784723103 314 / 578 The train loss 0.924412296148 315 / 578 The train loss 0.92547175248 316 / 578 The train loss 0.925892327099 317 / 578 The train loss 0.925825990031 318 / 578 The train loss 0.926876734708 319 / 578 The train loss 0.928485052035 320 / 578 The train loss 0.928756542201 321 / 578 The train loss 0.92851576411 322 / 578 The train loss 0.927860717665 323 / 578 The train loss 0.926540677277 324 / 578 The train loss 0.924966059602 325 / 578 The train loss 0.923692543942 326 / 578 The train loss 0.92173016645 327 / 578 The train loss 0.919717804987 328 / 578 The train loss 0.917494678575 329 / 578 The train loss 0.915256821661 330 / 578 The train loss 0.912838115453 331 / 578 The train loss 0.910273581181 332 / 578 The train loss 0.907729360376 333 / 578 The train loss 0.905710640053 334 / 578 The train loss 0.904568572222 335 / 578 The train loss 0.905546348464 336 / 578 The train loss 0.907544522036 337 / 578 The train loss 0.908586736227 338 / 578 The train loss 0.907276289681 339 / 578 The train loss 0.905163813327 340 / 578 The train loss 0.902975072208 341 / 578 The train loss 0.901547237204 342 / 578 The train loss 0.899456622635 343 / 578 The train loss 0.898200780653 344 / 578 The train loss 0.897123085963 345 / 578 The train loss 0.895931463807 346 / 578 The train loss 0.894706933549 347 / 578 The train loss 0.894199770084 348 / 578 The train loss 0.894387657508 349 / 578 The train loss 0.895056206957 350 / 578 The train loss 0.89530223938 351 / 578 The train loss 0.895510033389 352 / 578 The train loss 0.895467803623 353 / 578 The train loss 0.895430413821 354 / 578 The train loss 0.895781908563 355 / 578 The train loss 0.896206696164 356 / 578 The train loss 0.896839398242 357 / 578 The train loss 0.897360231926 358 / 578 The train loss 0.897698699658 359 / 578 The train loss 0.897789074912 360 / 578 The train loss 0.898094598183 361 / 578 The train loss 0.898380965469 362 / 578 The train loss 0.898370290556 363 / 578 The train loss 0.898110694691 364 / 578 The train loss 0.897364844885 365 / 578 The train loss 0.896534123547 366 / 578 The train loss 0.894405962441 367 / 578 The train loss 0.892214953229 368 / 578 The train loss 0.890106775292 369 / 578 The train loss 0.88792106695 370 / 578 The train loss 0.885707430763 371 / 578 The train loss 0.88356301797 372 / 578 The train loss 0.88136804947 373 / 578 The train loss 0.879197979361 374 / 578 The train loss 0.877002988897 375 / 578 The train loss 0.874900486281 376 / 578 The train loss 0.872783514681 377 / 578 The train loss 0.870688246528 378 / 578 The train loss 0.868474547786 379 / 578 The train loss 0.86652196468 380 / 578 The train loss 0.864502565916 381 / 578 The train loss 0.862644500776 382 / 578 The train loss 0.861127546164 383 / 578 The train loss 0.859543426619 384 / 578 The train loss 0.857664289186 385 / 578 The train loss 0.855687915931 386 / 578 The train loss 0.853786689591 387 / 578 The train loss 0.851871258529 388 / 578 The train loss 0.849968019642 389 / 578 The train loss 0.848254075623 390 / 578 The train loss 0.846339463767 391 / 578 The train loss 0.84442423427 392 / 578 The train loss 0.842478255817 393 / 578 The train loss 0.840555656161 394 / 578 The train loss 0.83867056091 395 / 578 The train loss 0.836735155292 396 / 578 The train loss 0.835012331699 397 / 578 The train loss 0.833593397075 398 / 578 The train loss 0.832306911018 399 / 578 The train loss 0.830861839604 400 / 578 The train loss 0.829572755937 401 / 578 The train loss 0.828179181458 402 / 578 The train loss 0.82680818003 403 / 578 The train loss 0.825751041896 404 / 578 The train loss 0.82481242755 405 / 578 The train loss 0.824026678171 406 / 578 The train loss 0.823098504576 407 / 578 The train loss 0.822298978796 408 / 578 The train loss 0.821439695289 409 / 578 The train loss 0.822516948076 410 / 578 The train loss 0.825355388933 411 / 578 The train loss 0.827409377226 412 / 578 The train loss 0.828066942345 413 / 578 The train loss 0.828681949427 414 / 578 The train loss 0.829707589576 415 / 578 The train loss 0.830590716776 416 / 578 The train loss 0.830307767673 417 / 578 The train loss 0.829714086434 418 / 578 The train loss 0.829318998123 419 / 578 The train loss 0.829390337775 420 / 578 The train loss 0.829392271542 421 / 578 The train loss 0.8292008068 422 / 578 The train loss 0.829082106121 423 / 578 The train loss 0.829189711305 424 / 578 The train loss 0.82908726654 425 / 578 The train loss 0.828842211909 426 / 578 The train loss 0.828308543426 427 / 578 The train loss 0.827509412529 428 / 578 The train loss 0.826548490168 429 / 578 The train loss 0.825311630217 430 / 578 The train loss 0.823894585756 431 / 578 The train loss 0.822314074259 432 / 578 The train loss 0.820594011702 433 / 578 The train loss 0.818797021536 434 / 578 The train loss 0.817052069605 435 / 578 The train loss 0.815307100648 436 / 578 The train loss 0.813542769342 437 / 578 The train loss 0.811898637392 438 / 578 The train loss 0.810564353969 439 / 578 The train loss 0.809320456932 440 / 578 The train loss 0.808147373499 441 / 578 The train loss 0.806797573094 442 / 578 The train loss 0.805295059761 443 / 578 The train loss 0.803983873837 444 / 578 The train loss 0.803457108914 445 / 578 The train loss 0.803370876736 446 / 578 The train loss 0.803567307888 447 / 578 The train loss 0.80390648068 448 / 578 The train loss 0.803888765505 449 / 578 The train loss 0.803627803034 450 / 578 The train loss 0.803151856073 451 / 578 The train loss 0.802649015408 452 / 578 The train loss 0.802256801561 453 / 578 The train loss 0.801975670314 454 / 578 The train loss 0.801541425263 455 / 578 The train loss 0.801015402401 456 / 578 The train loss 0.800343475329 457 / 578 The train loss 0.799640141322 458 / 578 The train loss 0.798867190712 459 / 578 The train loss 0.798018502392 460 / 578 The train loss 0.797459601327 461 / 578 The train loss 0.796798477471 462 / 578 The train loss 0.795983983784 463 / 578 The train loss 0.795485720851 464 / 578 The train loss 0.794715782507 465 / 578 The train loss 0.7939870812 466 / 578 The train loss 0.793049426366 467 / 578 The train loss 0.791974633174 468 / 578 The train loss 0.790992258586 469 / 578 The train loss 0.790335276877 470 / 578 The train loss 0.78984254777 471 / 578 The train loss 0.789321007251 472 / 578 The train loss 0.788927996444 473 / 578 The train loss 0.788400004588 474 / 578 The train loss 0.787632177281 475 / 578 The train loss 0.787049670855 476 / 578 The train loss 0.786458664339 477 / 578 The train loss 0.785840533947 478 / 578 The train loss 0.784812011986 479 / 578 The train loss 0.783582293018 480 / 578 The train loss 0.782321528415 481 / 578 The train loss 0.781111549261 482 / 578 The train loss 0.779895177467 483 / 578 The train loss 0.778542091457 484 / 578 The train loss 0.777379935969 485 / 578 The train loss 0.776094398955 486 / 578 The train loss 0.774898943319 487 / 578 The train loss 0.773677416261 488 / 578 The train loss 0.772515993848 489 / 578 The train loss 0.771291468567 490 / 578 The train loss 0.770039046427 491 / 578 The train loss 0.768744323539 492 / 578 The train loss 0.767486783343 493 / 578 The train loss 0.766243378037 494 / 578 The train loss 0.76502550744 495 / 578 The train loss 0.763960051514 496 / 578 The train loss 0.763149658856 497 / 578 The train loss 0.762412378703 498 / 578 The train loss 0.761745324313 499 / 578 The train loss 0.760974380178 500 / 578 The train loss 0.759968830325 501 / 578 The train loss 0.758834199731 502 / 578 The train loss 0.757792389584 503 / 578 The train loss 0.756897826533 504 / 578 The train loss 0.755940204967 505 / 578 The train loss 0.755018819005 506 / 578 The train loss 0.754048057438 507 / 578 The train loss 0.753064509305 508 / 578 The train loss 0.751926471787 509 / 578 The train loss 0.750675733313 510 / 578 The train loss 0.749419753889 511 / 578 The train loss 0.748191867432 512 / 578 The train loss 0.746937514552 513 / 578 The train loss 0.745654448395 514 / 578 The train loss 0.744408295421 515 / 578 The train loss 0.743210505969 516 / 578 The train loss 0.742090932094 517 / 578 The train loss 0.741044250106 518 / 578 The train loss 0.739970543433 519 / 578 The train loss 0.738813882525 520 / 578 The train loss 0.737780488992 521 / 578 The train loss 0.736758422621 522 / 578 The train loss 0.735576811586 523 / 578 The train loss 0.734505760076 524 / 578 The train loss 0.733598431579 525 / 578 The train loss 0.732775654984 526 / 578 The train loss 0.731864765534 527 / 578 The train loss 0.731063385298 528 / 578 The train loss 0.730340015277 529 / 578 The train loss 0.7295866597 530 / 578 The train loss 0.728886245072 531 / 578 The train loss 0.728077301219 532 / 578 The train loss 0.727296065252 533 / 578 The train loss 0.726500815364 534 / 578 The train loss 0.72557964764 535 / 578 The train loss 0.724497625559 536 / 578 The train loss 0.723416578296 537 / 578 The train loss 0.722468492421 538 / 578 The train loss 0.721582466058 539 / 578 The train loss 0.72069089548 540 / 578 The train loss 0.719930913927 541 / 578 The train loss 0.719059604284 542 / 578 The train loss 0.718275633376 543 / 578 The train loss 0.717353799883 544 / 578 The train loss 0.716763706268 545 / 578 The train loss 0.71634173223 546 / 578 The train loss 0.717109019114 547 / 578 The train loss 0.718558050426 548 / 578 The train loss 0.721306688642 549 / 578 The train loss 0.722884340412 550 / 578 The train loss 0.723908192942 551 / 578 The train loss 0.724700074061 552 / 578 The train loss 0.725243212351 553 / 578 The train loss 0.725872486069 554 / 578 The train loss 0.72654532315 555 / 578 The train loss 0.726663572771 556 / 578 The train loss 0.726654000895 557 / 578 The train loss 0.726729999372 558 / 578 The train loss 0.726943804739 559 / 578 The train loss 0.72741741757 560 / 578 The train loss 0.727844662798 561 / 578 The train loss 0.728137841777 562 / 578 The train loss 0.728534203454 563 / 578 The train loss 0.728643796524 564 / 578 The train loss 0.727837895717 565 / 578 The train loss 0.727063050675 566 / 578 The train loss 0.726234486313 567 / 578 The train loss 0.725342859213 568 / 578 The train loss 0.724493062327 569 / 578 The train loss 0.723619380157 570 / 578 The train loss 0.722805776218 571 / 578 The train loss 0.721926959476 572 / 578 The train loss 0.721225669145 573 / 578 The train loss 0.721403756814 574 / 578 The train loss 0.721547748692 575 / 578 The train loss 0.721232259708 576 / 578 The train loss 0.721158621231 577 / 578 The train loss 0.720938510063 578 / 578 The train loss 0.720923281478 Starting epoch 8 Validation: 1 / 30 The valid loss 0.463320881128 2 / 30 The valid loss 0.437654212117 3 / 30 The valid loss 0.416139225165 4 / 30 The valid loss 0.394769191742 5 / 30 The valid loss 0.375873804092 6 / 30 The valid loss 0.41640885671 7 / 30 The valid loss 0.530924831118 8 / 30 The valid loss 0.676041230559 9 / 30 The valid loss 0.824074493514 10 / 30 The valid loss 0.904589021206 11 / 30 The valid loss 0.922967108813 12 / 30 The valid loss 0.934605439504 13 / 30 The valid loss 0.943950524697 14 / 30 The valid loss 0.949057468346 15 / 30 The valid loss 0.931967373689 16 / 30 The valid loss 0.924649171531 17 / 30 The valid loss 0.937871603405 18 / 30 The valid loss 0.94780217939 19 / 30 The valid loss 0.948270929487 20 / 30 The valid loss 0.942089468241 21 / 30 The valid loss 0.928215296496 22 / 30 The valid loss 0.915294127031 23 / 30 The valid loss 0.90353581698 24 / 30 The valid loss 0.894812352955 25 / 30 The valid loss 0.88663244009 26 / 30 The valid loss 0.883165166928 27 / 30 The valid loss 0.872433735265 28 / 30 The valid loss 0.852887493159 29 / 30 The valid loss 0.830529499671 30 / 30 The valid loss 0.811760478218 Training 1 / 578 The train loss 0.586513876915 2 / 578 The train loss 0.480133146048 3 / 578 The train loss 0.422510027885 4 / 578 The train loss 0.390700951219 5 / 578 The train loss 0.383293050528 6 / 578 The train loss 0.356246280173 7 / 578 The train loss 0.328091225454 8 / 578 The train loss 0.299028171226 9 / 578 The train loss 0.273414613472 10 / 578 The train loss 0.256548983604 11 / 578 The train loss 0.245396460322 12 / 578 The train loss 0.237290914481 13 / 578 The train loss 0.233158045663 14 / 578 The train loss 0.238150228879 15 / 578 The train loss 0.237863724927 16 / 578 The train loss 0.23624628922 17 / 578 The train loss 0.23293097624 18 / 578 The train loss 0.228104000704 19 / 578 The train loss 0.221379753006 20 / 578 The train loss 0.213304861449 21 / 578 The train loss 0.207135540921 22 / 578 The train loss 0.201269600032 23 / 578 The train loss 0.195378412209 24 / 578 The train loss 0.190303828412 25 / 578 The train loss 0.186512588412 26 / 578 The train loss 0.182986081649 27 / 578 The train loss 0.181260738522 28 / 578 The train loss 0.178861940679 29 / 578 The train loss 0.175485855418 30 / 578 The train loss 0.173375096545 31 / 578 The train loss 0.170387053562 32 / 578 The train loss 0.167197500705 33 / 578 The train loss 0.166712480851 34 / 578 The train loss 0.167384413126 35 / 578 The train loss 0.171037391147 36 / 578 The train loss 0.174377462827 37 / 578 The train loss 0.177942347587 38 / 578 The train loss 0.180973588538 39 / 578 The train loss 0.182099273858 40 / 578 The train loss 0.181988895219 41 / 578 The train loss 0.181117125673 42 / 578 The train loss 0.180760338459 43 / 578 The train loss 0.182125881921 44 / 578 The train loss 0.183191994628 45 / 578 The train loss 0.184351378265 46 / 578 The train loss 0.183497414884 47 / 578 The train loss 0.182402059674 48 / 578 The train loss 0.18097901647 49 / 578 The train loss 0.179711371888 50 / 578 The train loss 0.180602988824 51 / 578 The train loss 0.182248857311 52 / 578 The train loss 0.183966339996 53 / 578 The train loss 0.184437968751 54 / 578 The train loss 0.18437756867 55 / 578 The train loss 0.184242452207 56 / 578 The train loss 0.183554005343 57 / 578 The train loss 0.184191186635 58 / 578 The train loss 0.186670440474 59 / 578 The train loss 0.190355283666 60 / 578 The train loss 0.197010994889 61 / 578 The train loss 0.206644045464 62 / 578 The train loss 0.215390291846 63 / 578 The train loss 0.222947688271 64 / 578 The train loss 0.22842768667 65 / 578 The train loss 0.231069180083 66 / 578 The train loss 0.231821266031 67 / 578 The train loss 0.232828228721 68 / 578 The train loss 0.233512826209 69 / 578 The train loss 0.234198458763 70 / 578 The train loss 0.236151310323 71 / 578 The train loss 0.240037203622 72 / 578 The train loss 0.24451609199 73 / 578 The train loss 0.247866792171 74 / 578 The train loss 0.251602580368 75 / 578 The train loss 0.255394268781 76 / 578 The train loss 0.25696471407 77 / 578 The train loss 0.257806264299 78 / 578 The train loss 0.259413123847 79 / 578 The train loss 0.261182893633 80 / 578 The train loss 0.262659741146 81 / 578 The train loss 0.263144548247 82 / 578 The train loss 0.263084894955 83 / 578 The train loss 0.264284715909 84 / 578 The train loss 0.265378465566 85 / 578 The train loss 0.266205281677 86 / 578 The train loss 0.265383639265 87 / 578 The train loss 0.265538680785 88 / 578 The train loss 0.265760217158 89 / 578 The train loss 0.268790510891 90 / 578 The train loss 0.27086851096 91 / 578 The train loss 0.269043063029 92 / 578 The train loss 0.267428169067 93 / 578 The train loss 0.265591070577 94 / 578 The train loss 0.263505393084 95 / 578 The train loss 0.262034518899 96 / 578 The train loss 0.26040465994 97 / 578 The train loss 0.259476874331 98 / 578 The train loss 0.258482985997 99 / 578 The train loss 0.258406493135 100 / 578 The train loss 0.258071932383 101 / 578 The train loss 0.257204403566 102 / 578 The train loss 0.256835729359 103 / 578 The train loss 0.256453544372 104 / 578 The train loss 0.255759710768 105 / 578 The train loss 0.255964044623 106 / 578 The train loss 0.255976641972 107 / 578 The train loss 0.256290025417 108 / 578 The train loss 0.255193932637 109 / 578 The train loss 0.254177278779 110 / 578 The train loss 0.253157463026 111 / 578 The train loss 0.252381669408 112 / 578 The train loss 0.252220706914 113 / 578 The train loss 0.251706083561 114 / 578 The train loss 0.252360496376 115 / 578 The train loss 0.255056633412 116 / 578 The train loss 0.257517059077 117 / 578 The train loss 0.25961532275 118 / 578 The train loss 0.263100779442 119 / 578 The train loss 0.266459794386 120 / 578 The train loss 0.270385663987 121 / 578 The train loss 0.274589011769 122 / 578 The train loss 0.279076398182 123 / 578 The train loss 0.286573706392 124 / 578 The train loss 0.295700678692 125 / 578 The train loss 0.305928422719 126 / 578 The train loss 0.315944738597 127 / 578 The train loss 0.325562304194 128 / 578 The train loss 0.335899778438 129 / 578 The train loss 0.346686378728 130 / 578 The train loss 0.358347412972 131 / 578 The train loss 0.367068969367 132 / 578 The train loss 0.372003838674 133 / 578 The train loss 0.374305480209 134 / 578 The train loss 0.376239397438 135 / 578 The train loss 0.379306894568 136 / 578 The train loss 0.383207396796 137 / 578 The train loss 0.38796330741 138 / 578 The train loss 0.390332835551 139 / 578 The train loss 0.392049096284 140 / 578 The train loss 0.392392112767 141 / 578 The train loss 0.391656699148 142 / 578 The train loss 0.391583427183 143 / 578 The train loss 0.393397507577 144 / 578 The train loss 0.395127246389 145 / 578 The train loss 0.396418764288 146 / 578 The train loss 0.397056634525 147 / 578 The train loss 0.397317444614 148 / 578 The train loss 0.39614697649 149 / 578 The train loss 0.395309197848 150 / 578 The train loss 0.394712951407 151 / 578 The train loss 0.392619206243 152 / 578 The train loss 0.390560792637 153 / 578 The train loss 0.388541705054 154 / 578 The train loss 0.386820913934 155 / 578 The train loss 0.386013762244 156 / 578 The train loss 0.385014688286 157 / 578 The train loss 0.384417739382 158 / 578 The train loss 0.383934964037 159 / 578 The train loss 0.382838752777 160 / 578 The train loss 0.381589330663 161 / 578 The train loss 0.380462956378 162 / 578 The train loss 0.379885543299 163 / 578 The train loss 0.379164234832 164 / 578 The train loss 0.377802155571 165 / 578 The train loss 0.37612568648 166 / 578 The train loss 0.375200888681 167 / 578 The train loss 0.374027438261 168 / 578 The train loss 0.377869957016 169 / 578 The train loss 0.385743236158 170 / 578 The train loss 0.394810140593 171 / 578 The train loss 0.403532598211 172 / 578 The train loss 0.409850649817 173 / 578 The train loss 0.417385566894 174 / 578 The train loss 0.424796704353 175 / 578 The train loss 0.431616710701 176 / 578 The train loss 0.438303743341 177 / 578 The train loss 0.442323729585 178 / 578 The train loss 0.444929923674 179 / 578 The train loss 0.448282325505 180 / 578 The train loss 0.450192058852 181 / 578 The train loss 0.452040431352 182 / 578 The train loss 0.453069495733 183 / 578 The train loss 0.454172582676 184 / 578 The train loss 0.457793625902 185 / 578 The train loss 0.462367499942 186 / 578 The train loss 0.470664680745 187 / 578 The train loss 0.481246658925 188 / 578 The train loss 0.492824245106 189 / 578 The train loss 0.503232716351 190 / 578 The train loss 0.517580531871 191 / 578 The train loss 0.530051835642 192 / 578 The train loss 0.540219287175 193 / 578 The train loss 0.548504733148 194 / 578 The train loss 0.554472625966 195 / 578 The train loss 0.558118848407 196 / 578 The train loss 0.562752290743 197 / 578 The train loss 0.569822180789 198 / 578 The train loss 0.573708535539 199 / 578 The train loss 0.578207899072 200 / 578 The train loss 0.583291341159 201 / 578 The train loss 0.592750594483 202 / 578 The train loss 0.60579373232 203 / 578 The train loss 0.616957242945 204 / 578 The train loss 0.625285768133 205 / 578 The train loss 0.631037838957 206 / 578 The train loss 0.63297209876 207 / 578 The train loss 0.63518748373 208 / 578 The train loss 0.641632791412 209 / 578 The train loss 0.650956114603 210 / 578 The train loss 0.657314985795 211 / 578 The train loss 0.662735188265 212 / 578 The train loss 0.668476641828 213 / 578 The train loss 0.673893382939 214 / 578 The train loss 0.67765467365 215 / 578 The train loss 0.681101298419 216 / 578 The train loss 0.683138643788 217 / 578 The train loss 0.684187670268 218 / 578 The train loss 0.68537269801 219 / 578 The train loss 0.687361814473 220 / 578 The train loss 0.689510135031 221 / 578 The train loss 0.693446288635 222 / 578 The train loss 0.69629919242 223 / 578 The train loss 0.699851518071 224 / 578 The train loss 0.703334248841 225 / 578 The train loss 0.709201041675 226 / 578 The train loss 0.715783690275 227 / 578 The train loss 0.722610758609 228 / 578 The train loss 0.727834024793 229 / 578 The train loss 0.733231358474 230 / 578 The train loss 0.738522506619 231 / 578 The train loss 0.743692487688 232 / 578 The train loss 0.748864295491 233 / 578 The train loss 0.754353450939 234 / 578 The train loss 0.760760042737 235 / 578 The train loss 0.768521747516 236 / 578 The train loss 0.773149018902 237 / 578 The train loss 0.774111994035 238 / 578 The train loss 0.773085712213 239 / 578 The train loss 0.772502353579 240 / 578 The train loss 0.772605175851 241 / 578 The train loss 0.773632347908 242 / 578 The train loss 0.774431464424 243 / 578 The train loss 0.774617372303 244 / 578 The train loss 0.775180038446 245 / 578 The train loss 0.776048507572 246 / 578 The train loss 0.776685837157 247 / 578 The train loss 0.774890296568 248 / 578 The train loss 0.773154830936 249 / 578 The train loss 0.771656364995 250 / 578 The train loss 0.771996043637 251 / 578 The train loss 0.771922012442 252 / 578 The train loss 0.772480202765 253 / 578 The train loss 0.773132384722 254 / 578 The train loss 0.774744827168 255 / 578 The train loss 0.776479272004 256 / 578 The train loss 0.780271352924 257 / 578 The train loss 0.783763401428 258 / 578 The train loss 0.785241238094 259 / 578 The train loss 0.786836982421 260 / 578 The train loss 0.787752000371 261 / 578 The train loss 0.788326821522 262 / 578 The train loss 0.788290855269 263 / 578 The train loss 0.787576065955 264 / 578 The train loss 0.787360820764 265 / 578 The train loss 0.787184089942 266 / 578 The train loss 0.787690591501 267 / 578 The train loss 0.788172765576 268 / 578 The train loss 0.789175620233 269 / 578 The train loss 0.791312749016 270 / 578 The train loss 0.793959090845 271 / 578 The train loss 0.796410854411 272 / 578 The train loss 0.796921076979 273 / 578 The train loss 0.796945839752 274 / 578 The train loss 0.796606633309 275 / 578 The train loss 0.796797429824 276 / 578 The train loss 0.797602025023 277 / 578 The train loss 0.797337949881 278 / 578 The train loss 0.797481796684 279 / 578 The train loss 0.798090813795 280 / 578 The train loss 0.798671585494 281 / 578 The train loss 0.798870164053 282 / 578 The train loss 0.798687320189 283 / 578 The train loss 0.798693873396 284 / 578 The train loss 0.798866101305 285 / 578 The train loss 0.798951236159 286 / 578 The train loss 0.799116385319 287 / 578 The train loss 0.799021371942 288 / 578 The train loss 0.799549995986 289 / 578 The train loss 0.799692812637 290 / 578 The train loss 0.800009148041 291 / 578 The train loss 0.800883337971 292 / 578 The train loss 0.802276037901 293 / 578 The train loss 0.802526530718 294 / 578 The train loss 0.803757205872 295 / 578 The train loss 0.803979778858 296 / 578 The train loss 0.805108388294 297 / 578 The train loss 0.805862493148 298 / 578 The train loss 0.808076525457 299 / 578 The train loss 0.809001430383 300 / 578 The train loss 0.810465189777 301 / 578 The train loss 0.811918366156 302 / 578 The train loss 0.814136344631 303 / 578 The train loss 0.815691048322 304 / 578 The train loss 0.816671340248 305 / 578 The train loss 0.818051445838 306 / 578 The train loss 0.819385538462 307 / 578 The train loss 0.820072124527 308 / 578 The train loss 0.821821126228 309 / 578 The train loss 0.822024712442 310 / 578 The train loss 0.821914955457 311 / 578 The train loss 0.823134900253 312 / 578 The train loss 0.82445856505 313 / 578 The train loss 0.82633113629 314 / 578 The train loss 0.82906492535 315 / 578 The train loss 0.829850962391 316 / 578 The train loss 0.830469010716 317 / 578 The train loss 0.831377273067 318 / 578 The train loss 0.832590456619 319 / 578 The train loss 0.832721537148 320 / 578 The train loss 0.832432506571 321 / 578 The train loss 0.831623700328 322 / 578 The train loss 0.83005224026 323 / 578 The train loss 0.828177902917 324 / 578 The train loss 0.826753504954 325 / 578 The train loss 0.825422741278 326 / 578 The train loss 0.823275212416 327 / 578 The train loss 0.821112850379 328 / 578 The train loss 0.818903008851 329 / 578 The train loss 0.816648781764 330 / 578 The train loss 0.814335011792 331 / 578 The train loss 0.812087063962 332 / 578 The train loss 0.809892082037 333 / 578 The train loss 0.808952760645 334 / 578 The train loss 0.809020006248 335 / 578 The train loss 0.811468061994 336 / 578 The train loss 0.814506730031 337 / 578 The train loss 0.814735382762 338 / 578 The train loss 0.813217149033 339 / 578 The train loss 0.811350196699 340 / 578 The train loss 0.809371626563 341 / 578 The train loss 0.807517701984 342 / 578 The train loss 0.805539874709 343 / 578 The train loss 0.804340678264 344 / 578 The train loss 0.803518984521 345 / 578 The train loss 0.802618840607 346 / 578 The train loss 0.801733141657 347 / 578 The train loss 0.801355468061 348 / 578 The train loss 0.801652220956 349 / 578 The train loss 0.802008560328 350 / 578 The train loss 0.802238477096 351 / 578 The train loss 0.802495203944 352 / 578 The train loss 0.802817797087 353 / 578 The train loss 0.80297508115 354 / 578 The train loss 0.803237256494 355 / 578 The train loss 0.803481942971 356 / 578 The train loss 0.804287253942 357 / 578 The train loss 0.804918298336 358 / 578 The train loss 0.80560917042 359 / 578 The train loss 0.805700941473 360 / 578 The train loss 0.806362541982 361 / 578 The train loss 0.80718201759 362 / 578 The train loss 0.807312496099 363 / 578 The train loss 0.806975547902 364 / 578 The train loss 0.806290167998 365 / 578 The train loss 0.805795553745 366 / 578 The train loss 0.803792072623 367 / 578 The train loss 0.801819944181 368 / 578 The train loss 0.799815674444 369 / 578 The train loss 0.797812541629 370 / 578 The train loss 0.795818922125 371 / 578 The train loss 0.793841724106 372 / 578 The train loss 0.791845305602 373 / 578 The train loss 0.789852135852 374 / 578 The train loss 0.787892640085 375 / 578 The train loss 0.785990695943 376 / 578 The train loss 0.784131560762 377 / 578 The train loss 0.78214424648 378 / 578 The train loss 0.780230427742 379 / 578 The train loss 0.778514471642 380 / 578 The train loss 0.776727464403 381 / 578 The train loss 0.775169437991 382 / 578 The train loss 0.773753748163 383 / 578 The train loss 0.772143910938 384 / 578 The train loss 0.770487082695 385 / 578 The train loss 0.768771290334 386 / 578 The train loss 0.767042271735 387 / 578 The train loss 0.765396347669 388 / 578 The train loss 0.763691181317 389 / 578 The train loss 0.762128602972 390 / 578 The train loss 0.760519162413 391 / 578 The train loss 0.758832662231 392 / 578 The train loss 0.757102503555 393 / 578 The train loss 0.755406785026 394 / 578 The train loss 0.753775939591 395 / 578 The train loss 0.75218830533 396 / 578 The train loss 0.750823355515 397 / 578 The train loss 0.749600917533 398 / 578 The train loss 0.748451662704 399 / 578 The train loss 0.747320313035 400 / 578 The train loss 0.74618093444 401 / 578 The train loss 0.745088358157 402 / 578 The train loss 0.744017513145 403 / 578 The train loss 0.742973686355 404 / 578 The train loss 0.74217200034 405 / 578 The train loss 0.741513705603 406 / 578 The train loss 0.740762181452 407 / 578 The train loss 0.739912625668 408 / 578 The train loss 0.739034644803 409 / 578 The train loss 0.740269550466 410 / 578 The train loss 0.742816846527 411 / 578 The train loss 0.744476296588 412 / 578 The train loss 0.744893396924 413 / 578 The train loss 0.745130823524 414 / 578 The train loss 0.746040162815 415 / 578 The train loss 0.746792343995 416 / 578 The train loss 0.746661013118 417 / 578 The train loss 0.746058850676 418 / 578 The train loss 0.745390448612 419 / 578 The train loss 0.745066058433 420 / 578 The train loss 0.744979423338 421 / 578 The train loss 0.744712031705 422 / 578 The train loss 0.744407946376 423 / 578 The train loss 0.744520785591 424 / 578 The train loss 0.744265493269 425 / 578 The train loss 0.743824719454 426 / 578 The train loss 0.743446543795 427 / 578 The train loss 0.742616694045 428 / 578 The train loss 0.741644653396 429 / 578 The train loss 0.740575592748 430 / 578 The train loss 0.739232049865 431 / 578 The train loss 0.73777329111 432 / 578 The train loss 0.736256895432 433 / 578 The train loss 0.734670852877 434 / 578 The train loss 0.733282622371 435 / 578 The train loss 0.731932393359 436 / 578 The train loss 0.730553339748 437 / 578 The train loss 0.729248097776 438 / 578 The train loss 0.728121087389 439 / 578 The train loss 0.727059925836 440 / 578 The train loss 0.725863369681 441 / 578 The train loss 0.724590520157 442 / 578 The train loss 0.723372814368 443 / 578 The train loss 0.72236073208 444 / 578 The train loss 0.721962546468 445 / 578 The train loss 0.72198699367 446 / 578 The train loss 0.722137162796 447 / 578 The train loss 0.722039453694 448 / 578 The train loss 0.721802776438 449 / 578 The train loss 0.721509127971 450 / 578 The train loss 0.720915373655 451 / 578 The train loss 0.720292726164 452 / 578 The train loss 0.719791681193 453 / 578 The train loss 0.719479905413 454 / 578 The train loss 0.718839196472 455 / 578 The train loss 0.718235070193 456 / 578 The train loss 0.71737870132 457 / 578 The train loss 0.716628351043 458 / 578 The train loss 0.716027132052 459 / 578 The train loss 0.715258926567 460 / 578 The train loss 0.714652804277 461 / 578 The train loss 0.714142597573 462 / 578 The train loss 0.713643737129 463 / 578 The train loss 0.713402073492 464 / 578 The train loss 0.713205610844 465 / 578 The train loss 0.71292955897 466 / 578 The train loss 0.712374278364 467 / 578 The train loss 0.711711273164 468 / 578 The train loss 0.711178404359 469 / 578 The train loss 0.71080033493 470 / 578 The train loss 0.710529790747 471 / 578 The train loss 0.710218267543 472 / 578 The train loss 0.70975214883 473 / 578 The train loss 0.709103283077 474 / 578 The train loss 0.708250232096 475 / 578 The train loss 0.70741001853 476 / 578 The train loss 0.706610206006 477 / 578 The train loss 0.705729490767 478 / 578 The train loss 0.704584637657 479 / 578 The train loss 0.703311675751 480 / 578 The train loss 0.702119458754 481 / 578 The train loss 0.701167847233 482 / 578 The train loss 0.700113542818 483 / 578 The train loss 0.699106137397 484 / 578 The train loss 0.6980753949 485 / 578 The train loss 0.696960792875 486 / 578 The train loss 0.695767668072 487 / 578 The train loss 0.694570190325 488 / 578 The train loss 0.693615113515 489 / 578 The train loss 0.692533499464 490 / 578 The train loss 0.691586231843 491 / 578 The train loss 0.69068655987 492 / 578 The train loss 0.689981143672 493 / 578 The train loss 0.689037431591 494 / 578 The train loss 0.688041008639 495 / 578 The train loss 0.686922094008 496 / 578 The train loss 0.686035715847 497 / 578 The train loss 0.685237805331 498 / 578 The train loss 0.6843908266 499 / 578 The train loss 0.68370664116 500 / 578 The train loss 0.68295317211 501 / 578 The train loss 0.681964665317 502 / 578 The train loss 0.681148726886 503 / 578 The train loss 0.680398137525 504 / 578 The train loss 0.679570374766 505 / 578 The train loss 0.678699431052 506 / 578 The train loss 0.677859820359 507 / 578 The train loss 0.676938035953 508 / 578 The train loss 0.675972175872 509 / 578 The train loss 0.674850276263 510 / 578 The train loss 0.673743844434 511 / 578 The train loss 0.672690295069 512 / 578 The train loss 0.671536955248 513 / 578 The train loss 0.670434520164 514 / 578 The train loss 0.669457209822 515 / 578 The train loss 0.66838733122 516 / 578 The train loss 0.667298991024 517 / 578 The train loss 0.666358223498 518 / 578 The train loss 0.665446402016 519 / 578 The train loss 0.664423105797 520 / 578 The train loss 0.663400703088 521 / 578 The train loss 0.662389123415 522 / 578 The train loss 0.661273265557 523 / 578 The train loss 0.66020917916 524 / 578 The train loss 0.659162244191 525 / 578 The train loss 0.658231942859 526 / 578 The train loss 0.657283433525 527 / 578 The train loss 0.656456790773 528 / 578 The train loss 0.65584801632 529 / 578 The train loss 0.655159757055 530 / 578 The train loss 0.654575726931 531 / 578 The train loss 0.653876318524 532 / 578 The train loss 0.653114741185 533 / 578 The train loss 0.652336710005 534 / 578 The train loss 0.651442263017 535 / 578 The train loss 0.650492470656 536 / 578 The train loss 0.649636562031 537 / 578 The train loss 0.648822209695 538 / 578 The train loss 0.64819253162 539 / 578 The train loss 0.647493859674 540 / 578 The train loss 0.646588603938 541 / 578 The train loss 0.645771078097 542 / 578 The train loss 0.64498397222 543 / 578 The train loss 0.644266150912 544 / 578 The train loss 0.643662423916 545 / 578 The train loss 0.643548079621 546 / 578 The train loss 0.644795539212 547 / 578 The train loss 0.646561905966 548 / 578 The train loss 0.649987473139 549 / 578 The train loss 0.651801219003 550 / 578 The train loss 0.652852933657 551 / 578 The train loss 0.653706703808 552 / 578 The train loss 0.654113997294 553 / 578 The train loss 0.654609073816 554 / 578 The train loss 0.655277280419 555 / 578 The train loss 0.65557807784 556 / 578 The train loss 0.655706068307 557 / 578 The train loss 0.656183349841 558 / 578 The train loss 0.656391092025 559 / 578 The train loss 0.656721359962 560 / 578 The train loss 0.657083359075 561 / 578 The train loss 0.657461609253 562 / 578 The train loss 0.657966115436 563 / 578 The train loss 0.658042274541 564 / 578 The train loss 0.65736759415 565 / 578 The train loss 0.656784230926 566 / 578 The train loss 0.65618820576 567 / 578 The train loss 0.655511236571 568 / 578 The train loss 0.654732884032 569 / 578 The train loss 0.653926970807 570 / 578 The train loss 0.653155207379 571 / 578 The train loss 0.652561011077 572 / 578 The train loss 0.652080291329 573 / 578 The train loss 0.652267961921 574 / 578 The train loss 0.652399083262 575 / 578 The train loss 0.652143877077 576 / 578 The train loss 0.652041648004 577 / 578 The train loss 0.651980067153 578 / 578 The train loss 0.651876786389 Starting epoch 9 Validation: 1 / 30 The valid loss 0.449623316526 2 / 30 The valid loss 0.423004865646 3 / 30 The valid loss 0.396767665943 4 / 30 The valid loss 0.359865427017 5 / 30 The valid loss 0.3308940202 6 / 30 The valid loss 0.362461787959 7 / 30 The valid loss 0.467353550451 8 / 30 The valid loss 0.604988815263 9 / 30 The valid loss 0.747355542249 10 / 30 The valid loss 0.824789525568 11 / 30 The valid loss 0.840900032358 12 / 30 The valid loss 0.850778675328 13 / 30 The valid loss 0.858655912372 14 / 30 The valid loss 0.862202956208 15 / 30 The valid loss 0.845962024728 16 / 30 The valid loss 0.839240537025 17 / 30 The valid loss 0.852555451148 18 / 30 The valid loss 0.863202678661 19 / 30 The valid loss 0.863749633494 20 / 30 The valid loss 0.857808787376 21 / 30 The valid loss 0.843305202467 22 / 30 The valid loss 0.829730912366 23 / 30 The valid loss 0.817935550342 24 / 30 The valid loss 0.809358761335 25 / 30 The valid loss 0.801181922555 26 / 30 The valid loss 0.79697453116 27 / 30 The valid loss 0.785773557093 28 / 30 The valid loss 0.767113844731 29 / 30 The valid loss 0.747124948378 30 / 30 The valid loss 0.730864402652 Validation MSE(val_loss): 15.5000451153 Test MSE(test_loss): 4.81319693507 Training 1 / 578 The train loss 0.510433375835 2 / 578 The train loss 0.42087328434 3 / 578 The train loss 0.458673000336 4 / 578 The train loss 0.49024066329 5 / 578 The train loss 0.487226843834 6 / 578 The train loss 0.431470602751 7 / 578 The train loss 0.389931646841 8 / 578 The train loss 0.354504168034 9 / 578 The train loss 0.329666503602 10 / 578 The train loss 0.315461188555 11 / 578 The train loss 0.300094673579 12 / 578 The train loss 0.287275829663 13 / 578 The train loss 0.280046641827 14 / 578 The train loss 0.281216091343 15 / 578 The train loss 0.281915360689 16 / 578 The train loss 0.284386508167 17 / 578 The train loss 0.281695151154 18 / 578 The train loss 0.273207617303 19 / 578 The train loss 0.264519954198 20 / 578 The train loss 0.255606216565 21 / 578 The train loss 0.246026521992 22 / 578 The train loss 0.237452576106 23 / 578 The train loss 0.22814243133 24 / 578 The train loss 0.220514996676 25 / 578 The train loss 0.215086414292 26 / 578 The train loss 0.210711836027 27 / 578 The train loss 0.206130543034 28 / 578 The train loss 0.202804525316 29 / 578 The train loss 0.198830639301 30 / 578 The train loss 0.195683797511 31 / 578 The train loss 0.19245958887 32 / 578 The train loss 0.189610553498 33 / 578 The train loss 0.187194311314 34 / 578 The train loss 0.186899974997 35 / 578 The train loss 0.188477308037 36 / 578 The train loss 0.191232297621 37 / 578 The train loss 0.193781541459 38 / 578 The train loss 0.198529536444 39 / 578 The train loss 0.198994752354 40 / 578 The train loss 0.197925853403 41 / 578 The train loss 0.195531352491 42 / 578 The train loss 0.193170671174 43 / 578 The train loss 0.193751280608 44 / 578 The train loss 0.194048681254 45 / 578 The train loss 0.193200467899 46 / 578 The train loss 0.1912819418 47 / 578 The train loss 0.189148912603 48 / 578 The train loss 0.187691533046 49 / 578 The train loss 0.18646790456 50 / 578 The train loss 0.185857457556 51 / 578 The train loss 0.186364196131 52 / 578 The train loss 0.187096482608 53 / 578 The train loss 0.186901630131 54 / 578 The train loss 0.185968250623 55 / 578 The train loss 0.184914364361 56 / 578 The train loss 0.183618446579 57 / 578 The train loss 0.182418658447 58 / 578 The train loss 0.182778671288 59 / 578 The train loss 0.184086155494 60 / 578 The train loss 0.188251919455 61 / 578 The train loss 0.194382569737 62 / 578 The train loss 0.199744808968 63 / 578 The train loss 0.203825445314 64 / 578 The train loss 0.205328742421 65 / 578 The train loss 0.205863208868 66 / 578 The train loss 0.204776988252 67 / 578 The train loss 0.203175900726 68 / 578 The train loss 0.201878285063 69 / 578 The train loss 0.200625814309 70 / 578 The train loss 0.200543512111 71 / 578 The train loss 0.201482978007 72 / 578 The train loss 0.203587255053 73 / 578 The train loss 0.204953682673 74 / 578 The train loss 0.206967394824 75 / 578 The train loss 0.20917222095 76 / 578 The train loss 0.210386662033 77 / 578 The train loss 0.210210459694 78 / 578 The train loss 0.209609259517 79 / 578 The train loss 0.211235904359 80 / 578 The train loss 0.212093106774 81 / 578 The train loss 0.21284381928 82 / 578 The train loss 0.214835132154 83 / 578 The train loss 0.217543901202 84 / 578 The train loss 0.221305168566 85 / 578 The train loss 0.224361990589 86 / 578 The train loss 0.226686046675 87 / 578 The train loss 0.230223517415 88 / 578 The train loss 0.232068793785 89 / 578 The train loss 0.238212438551 90 / 578 The train loss 0.242760336171 91 / 578 The train loss 0.242135651921 92 / 578 The train loss 0.241807260206 93 / 578 The train loss 0.240927133128 94 / 578 The train loss 0.239339253627 95 / 578 The train loss 0.23799429977 96 / 578 The train loss 0.237011609618 97 / 578 The train loss 0.235835616141 98 / 578 The train loss 0.235542559825 99 / 578 The train loss 0.236128221275 100 / 578 The train loss 0.23540312456 101 / 578 The train loss 0.235351441539 102 / 578 The train loss 0.234874462252 103 / 578 The train loss 0.234232121174 104 / 578 The train loss 0.233123258992 105 / 578 The train loss 0.233168971343 106 / 578 The train loss 0.233286481509 107 / 578 The train loss 0.233790553149 108 / 578 The train loss 0.234523817983 109 / 578 The train loss 0.235109769509 110 / 578 The train loss 0.234616957283 111 / 578 The train loss 0.233622684286 112 / 578 The train loss 0.232672032917 113 / 578 The train loss 0.232877123029 114 / 578 The train loss 0.234299636086 115 / 578 The train loss 0.237213166269 116 / 578 The train loss 0.24045575887 117 / 578 The train loss 0.242531277319 118 / 578 The train loss 0.247271800587 119 / 578 The train loss 0.252427005107 120 / 578 The train loss 0.256742720446 121 / 578 The train loss 0.261151326728 122 / 578 The train loss 0.265987305505 123 / 578 The train loss 0.273328045656 124 / 578 The train loss 0.282818929938 125 / 578 The train loss 0.292427038327 126 / 578 The train loss 0.301305627199 127 / 578 The train loss 0.310832010611 128 / 578 The train loss 0.32006791899 129 / 578 The train loss 0.331023803183 130 / 578 The train loss 0.342500610664 131 / 578 The train loss 0.350111817714 132 / 578 The train loss 0.353781199492 133 / 578 The train loss 0.355110587608 134 / 578 The train loss 0.356774754169 135 / 578 The train loss 0.359935532359 136 / 578 The train loss 0.364897945033 137 / 578 The train loss 0.369860822824 138 / 578 The train loss 0.372374784809 139 / 578 The train loss 0.373877061477 140 / 578 The train loss 0.374636118447 141 / 578 The train loss 0.375368473285 142 / 578 The train loss 0.376122587492 143 / 578 The train loss 0.3781989987 144 / 578 The train loss 0.38032072939 145 / 578 The train loss 0.381713929169 146 / 578 The train loss 0.382120455896 147 / 578 The train loss 0.382432298709 148 / 578 The train loss 0.382164600242 149 / 578 The train loss 0.382056175575 150 / 578 The train loss 0.381652991287 151 / 578 The train loss 0.379569224039 152 / 578 The train loss 0.377805493686 153 / 578 The train loss 0.376060703331 154 / 578 The train loss 0.374252407207 155 / 578 The train loss 0.372621667109 156 / 578 The train loss 0.371220714019 157 / 578 The train loss 0.370159210495 158 / 578 The train loss 0.369730787198 159 / 578 The train loss 0.368982763047 160 / 578 The train loss 0.368323390943 161 / 578 The train loss 0.36750182455 162 / 578 The train loss 0.366319758484 163 / 578 The train loss 0.365624152638 164 / 578 The train loss 0.36432662889 165 / 578 The train loss 0.363036635981 166 / 578 The train loss 0.361786688229 167 / 578 The train loss 0.360452847221 168 / 578 The train loss 0.363891607605 169 / 578 The train loss 0.371142506963 170 / 578 The train loss 0.379070510244 171 / 578 The train loss 0.387292117066 172 / 578 The train loss 0.393084703154 173 / 578 The train loss 0.397869178681 174 / 578 The train loss 0.402863847466 175 / 578 The train loss 0.408469534272 176 / 578 The train loss 0.414106538699 177 / 578 The train loss 0.417557200143 178 / 578 The train loss 0.419379307993 179 / 578 The train loss 0.422585397803 180 / 578 The train loss 0.425244429718 181 / 578 The train loss 0.426567726053 182 / 578 The train loss 0.426922201466 183 / 578 The train loss 0.427773000774 184 / 578 The train loss 0.431788810579 185 / 578 The train loss 0.436254579322 186 / 578 The train loss 0.441416236031 187 / 578 The train loss 0.449579868097 188 / 578 The train loss 0.458631266315 189 / 578 The train loss 0.46814890199 190 / 578 The train loss 0.480444015337 191 / 578 The train loss 0.490128092126 192 / 578 The train loss 0.498789766813 193 / 578 The train loss 0.505693338088 194 / 578 The train loss 0.51008695918 195 / 578 The train loss 0.513157523605 196 / 578 The train loss 0.516214724716 197 / 578 The train loss 0.521917090208 198 / 578 The train loss 0.525404370804 199 / 578 The train loss 0.529449489258 200 / 578 The train loss 0.534043375263 201 / 578 The train loss 0.541532910295 202 / 578 The train loss 0.55279489792 203 / 578 The train loss 0.562406693782 204 / 578 The train loss 0.56883281704 205 / 578 The train loss 0.572506544076 206 / 578 The train loss 0.57321671694 207 / 578 The train loss 0.574222024809 208 / 578 The train loss 0.578677304958 209 / 578 The train loss 0.586797995154 210 / 578 The train loss 0.591616866427 211 / 578 The train loss 0.595509253698 212 / 578 The train loss 0.598624520434 213 / 578 The train loss 0.600669498464 214 / 578 The train loss 0.601694090631 215 / 578 The train loss 0.602335683698 216 / 578 The train loss 0.602189743693 217 / 578 The train loss 0.601463618238 218 / 578 The train loss 0.601047284158 219 / 578 The train loss 0.601076242731 220 / 578 The train loss 0.600795490235 221 / 578 The train loss 0.601448235612 222 / 578 The train loss 0.601648998027 223 / 578 The train loss 0.601858453508 224 / 578 The train loss 0.603646967169 225 / 578 The train loss 0.6071806517 226 / 578 The train loss 0.610666082636 227 / 578 The train loss 0.613166344463 228 / 578 The train loss 0.616847367802 229 / 578 The train loss 0.619904171049 230 / 578 The train loss 0.622046323064 231 / 578 The train loss 0.624662352468 232 / 578 The train loss 0.627409995253 233 / 578 The train loss 0.629086393791 234 / 578 The train loss 0.631699055823 235 / 578 The train loss 0.636830213357 236 / 578 The train loss 0.639076301236 237 / 578 The train loss 0.638451212703 238 / 578 The train loss 0.63660142152 239 / 578 The train loss 0.634736219343 240 / 578 The train loss 0.632568492872 241 / 578 The train loss 0.630440770567 242 / 578 The train loss 0.628723856562 243 / 578 The train loss 0.627693111368 244 / 578 The train loss 0.626482387157 245 / 578 The train loss 0.62624594129 246 / 578 The train loss 0.626570598964 247 / 578 The train loss 0.625144573939 248 / 578 The train loss 0.62346269158 249 / 578 The train loss 0.621960586131 250 / 578 The train loss 0.621243644036 251 / 578 The train loss 0.620679653892 252 / 578 The train loss 0.621494694044 253 / 578 The train loss 0.622437848087 254 / 578 The train loss 0.62413096678 255 / 578 The train loss 0.627080732387 256 / 578 The train loss 0.631121371094 257 / 578 The train loss 0.63291701423 258 / 578 The train loss 0.634711849749 259 / 578 The train loss 0.635361957885 260 / 578 The train loss 0.63609156211 261 / 578 The train loss 0.636284779921 262 / 578 The train loss 0.636195093315 263 / 578 The train loss 0.635308828339 264 / 578 The train loss 0.635249917402 265 / 578 The train loss 0.635883685026 266 / 578 The train loss 0.636721535432 267 / 578 The train loss 0.636732743174 268 / 578 The train loss 0.637116854906 269 / 578 The train loss 0.639018079307 270 / 578 The train loss 0.641762966725 271 / 578 The train loss 0.644554486718 272 / 578 The train loss 0.646282755859 273 / 578 The train loss 0.645804417364 274 / 578 The train loss 0.64549306963 275 / 578 The train loss 0.645032927265 276 / 578 The train loss 0.645143074328 277 / 578 The train loss 0.64423759086 278 / 578 The train loss 0.64361062199 279 / 578 The train loss 0.643006237744 280 / 578 The train loss 0.642342281268 281 / 578 The train loss 0.641290950639 282 / 578 The train loss 0.640146151958 283 / 578 The train loss 0.639180504635 284 / 578 The train loss 0.638444652892 285 / 578 The train loss 0.638158953223 286 / 578 The train loss 0.638631963471 287 / 578 The train loss 0.639143857008 288 / 578 The train loss 0.640206975895 289 / 578 The train loss 0.641247331097 290 / 578 The train loss 0.643028457855 291 / 578 The train loss 0.644957866078 292 / 578 The train loss 0.646925255484 293 / 578 The train loss 0.648808638422 294 / 578 The train loss 0.649597601452 295 / 578 The train loss 0.649908652176 296 / 578 The train loss 0.650645649344 297 / 578 The train loss 0.651480789368 298 / 578 The train loss 0.651781492417 299 / 578 The train loss 0.652668479923 300 / 578 The train loss 0.653795860838 301 / 578 The train loss 0.655646020225 302 / 578 The train loss 0.657248814208 303 / 578 The train loss 0.659313012105 304 / 578 The train loss 0.661348650557 305 / 578 The train loss 0.662502233153 306 / 578 The train loss 0.66373814889 307 / 578 The train loss 0.66456309129 308 / 578 The train loss 0.665717739207 309 / 578 The train loss 0.666660358249 310 / 578 The train loss 0.667532783392 311 / 578 The train loss 0.67020754476 312 / 578 The train loss 0.673293513968 313 / 578 The train loss 0.675775209775 314 / 578 The train loss 0.679209905285 315 / 578 The train loss 0.681817232078 316 / 578 The train loss 0.684050739407 317 / 578 The train loss 0.686271700791 318 / 578 The train loss 0.687490522328 319 / 578 The train loss 0.688418955294 320 / 578 The train loss 0.688356328412 321 / 578 The train loss 0.687835943876 322 / 578 The train loss 0.686531200016 323 / 578 The train loss 0.685562264468 324 / 578 The train loss 0.684502503783 325 / 578 The train loss 0.68324374752 326 / 578 The train loss 0.681601724907 327 / 578 The train loss 0.679857445585 328 / 578 The train loss 0.678447692206 329 / 578 The train loss 0.677253464962 330 / 578 The train loss 0.67760345302 331 / 578 The train loss 0.679535373623 332 / 578 The train loss 0.682300442476 333 / 578 The train loss 0.685213362295 334 / 578 The train loss 0.689195445003 335 / 578 The train loss 0.696990697581 336 / 578 The train loss 0.705870972309 337 / 578 The train loss 0.708661205144 338 / 578 The train loss 0.70795576793 339 / 578 The train loss 0.706277102699 340 / 578 The train loss 0.704789504787 341 / 578 The train loss 0.703312108839 342 / 578 The train loss 0.702106865771 343 / 578 The train loss 0.701619005135 344 / 578 The train loss 0.701125496246 345 / 578 The train loss 0.700705792318 346 / 578 The train loss 0.699787971585 347 / 578 The train loss 0.699294265633 348 / 578 The train loss 0.699404797147 349 / 578 The train loss 0.700507985983 350 / 578 The train loss 0.70181789652 351 / 578 The train loss 0.703042404079 352 / 578 The train loss 0.7037892609 353 / 578 The train loss 0.703974209984 354 / 578 The train loss 0.704704334659 355 / 578 The train loss 0.706001985057 356 / 578 The train loss 0.707697855439 357 / 578 The train loss 0.709264538353 358 / 578 The train loss 0.710462621288 359 / 578 The train loss 0.711697891381 360 / 578 The train loss 0.712795569649 361 / 578 The train loss 0.713750718523 362 / 578 The train loss 0.714160296783 363 / 578 The train loss 0.714047168718 364 / 578 The train loss 0.713477912006 365 / 578 The train loss 0.713389521414 366 / 578 The train loss 0.711674900822 367 / 578 The train loss 0.710251445306 368 / 578 The train loss 0.708815418634 369 / 578 The train loss 0.707195404912 370 / 578 The train loss 0.705581494043 371 / 578 The train loss 0.703907308437 372 / 578 The train loss 0.702283890102 373 / 578 The train loss 0.700672842947 374 / 578 The train loss 0.698916033194 375 / 578 The train loss 0.697202933128 376 / 578 The train loss 0.695579687881 377 / 578 The train loss 0.69394390241 378 / 578 The train loss 0.692240845644 379 / 578 The train loss 0.690802170386 380 / 578 The train loss 0.689293191401 381 / 578 The train loss 0.688101486442 382 / 578 The train loss 0.687161695129 383 / 578 The train loss 0.686015436465 384 / 578 The train loss 0.684611961609 385 / 578 The train loss 0.683105517087 386 / 578 The train loss 0.681536923394 387 / 578 The train loss 0.680003506707 388 / 578 The train loss 0.678639278099 389 / 578 The train loss 0.67721184266 390 / 578 The train loss 0.675685381694 391 / 578 The train loss 0.674174332553 392 / 578 The train loss 0.672664184042 393 / 578 The train loss 0.671146451493 394 / 578 The train loss 0.669642514001 395 / 578 The train loss 0.668106241859 396 / 578 The train loss 0.667019354325 397 / 578 The train loss 0.66614440434 398 / 578 The train loss 0.66532293439 399 / 578 The train loss 0.664384584281 400 / 578 The train loss 0.663486348367 401 / 578 The train loss 0.662500516263 402 / 578 The train loss 0.661450115505 403 / 578 The train loss 0.660622769098 404 / 578 The train loss 0.659702833608 405 / 578 The train loss 0.658982952118 406 / 578 The train loss 0.658091113389 407 / 578 The train loss 0.657073339219 408 / 578 The train loss 0.656104025865 409 / 578 The train loss 0.657312600106 410 / 578 The train loss 0.659916153164 411 / 578 The train loss 0.661507430135 412 / 578 The train loss 0.661761136142 413 / 578 The train loss 0.661694878361 414 / 578 The train loss 0.661816826887 415 / 578 The train loss 0.661840031642 416 / 578 The train loss 0.661301418758 417 / 578 The train loss 0.66065912729 418 / 578 The train loss 0.659931921694 419 / 578 The train loss 0.658993079703 420 / 578 The train loss 0.658377277545 421 / 578 The train loss 0.657769648712 422 / 578 The train loss 0.657373853112 423 / 578 The train loss 0.657362503411 424 / 578 The train loss 0.657311118485 425 / 578 The train loss 0.6570287618 426 / 578 The train loss 0.656848063399 427 / 578 The train loss 0.656344655714 428 / 578 The train loss 0.655614128924 429 / 578 The train loss 0.654557367457 430 / 578 The train loss 0.653120570288 431 / 578 The train loss 0.651826797693 432 / 578 The train loss 0.650623072064 433 / 578 The train loss 0.649330795831 434 / 578 The train loss 0.648199143416 435 / 578 The train loss 0.647162789128 436 / 578 The train loss 0.645970828408 437 / 578 The train loss 0.644689041776 438 / 578 The train loss 0.643682041095 439 / 578 The train loss 0.64269825949 440 / 578 The train loss 0.641776948122 441 / 578 The train loss 0.641007007152 442 / 578 The train loss 0.640322494599 443 / 578 The train loss 0.639586384192 444 / 578 The train loss 0.639302211541 445 / 578 The train loss 0.639238365953 446 / 578 The train loss 0.63931122446 447 / 578 The train loss 0.639290849472 448 / 578 The train loss 0.639326732976 449 / 578 The train loss 0.639047185108 450 / 578 The train loss 0.638362275465 451 / 578 The train loss 0.637635913998 452 / 578 The train loss 0.637088457011 453 / 578 The train loss 0.636717861428 454 / 578 The train loss 0.636132421313 455 / 578 The train loss 0.635582453704 456 / 578 The train loss 0.635008076275 457 / 578 The train loss 0.634435207014 458 / 578 The train loss 0.633952276972 459 / 578 The train loss 0.633400652245 460 / 578 The train loss 0.632633676967 461 / 578 The train loss 0.632166105183 462 / 578 The train loss 0.631979516933 463 / 578 The train loss 0.632110765735 464 / 578 The train loss 0.632910552053 465 / 578 The train loss 0.633433306446 466 / 578 The train loss 0.633053264493 467 / 578 The train loss 0.632705076959 468 / 578 The train loss 0.632322183313 469 / 578 The train loss 0.632031714266 470 / 578 The train loss 0.631652307673 471 / 578 The train loss 0.631306878673 472 / 578 The train loss 0.631041934781 473 / 578 The train loss 0.630688142043 474 / 578 The train loss 0.629912503068 475 / 578 The train loss 0.629159621036 476 / 578 The train loss 0.628301439294 477 / 578 The train loss 0.627306618035 478 / 578 The train loss 0.62630482539 479 / 578 The train loss 0.625527890562 480 / 578 The train loss 0.62463144114 481 / 578 The train loss 0.62386916792 482 / 578 The train loss 0.623052488135 483 / 578 The train loss 0.622338680473 484 / 578 The train loss 0.621426555986 485 / 578 The train loss 0.620486739712 486 / 578 The train loss 0.619478737748 487 / 578 The train loss 0.618589702669 488 / 578 The train loss 0.617914000835 489 / 578 The train loss 0.617176989261 490 / 578 The train loss 0.616549217431 491 / 578 The train loss 0.615906464019 492 / 578 The train loss 0.61521956431 493 / 578 The train loss 0.614430319469 494 / 578 The train loss 0.613625546069 495 / 578 The train loss 0.612724436086 496 / 578 The train loss 0.612142576119 497 / 578 The train loss 0.611667505606 498 / 578 The train loss 0.611060009817 499 / 578 The train loss 0.61060571859 500 / 578 The train loss 0.609795745227 501 / 578 The train loss 0.608989828067 502 / 578 The train loss 0.608183233925 503 / 578 The train loss 0.60745540863 504 / 578 The train loss 0.606903779753 505 / 578 The train loss 0.606106590933 506 / 578 The train loss 0.605425109862 507 / 578 The train loss 0.604560972381 508 / 578 The train loss 0.603590196457 509 / 578 The train loss 0.60261013712 510 / 578 The train loss 0.601559780819 511 / 578 The train loss 0.600522416551 512 / 578 The train loss 0.599525384026 513 / 578 The train loss 0.598544019636 514 / 578 The train loss 0.597690328481 515 / 578 The train loss 0.596893894488 516 / 578 The train loss 0.596127221826 517 / 578 The train loss 0.59542727109 518 / 578 The train loss 0.5944733386 519 / 578 The train loss 0.593495236473 520 / 578 The train loss 0.592614136769 521 / 578 The train loss 0.591774780085 522 / 578 The train loss 0.590817181119 523 / 578 The train loss 0.589909153408 524 / 578 The train loss 0.589030468243 525 / 578 The train loss 0.588253411328 526 / 578 The train loss 0.587602840563 527 / 578 The train loss 0.587011722042 528 / 578 The train loss 0.586411587714 529 / 578 The train loss 0.585953978815 530 / 578 The train loss 0.585377046399 531 / 578 The train loss 0.584845926399 532 / 578 The train loss 0.584311370216 533 / 578 The train loss 0.58369407504 534 / 578 The train loss 0.582863540954 535 / 578 The train loss 0.58207804593 536 / 578 The train loss 0.581283783947 537 / 578 The train loss 0.580541971832 538 / 578 The train loss 0.579931772487 539 / 578 The train loss 0.579255554213 540 / 578 The train loss 0.578662324245 541 / 578 The train loss 0.577972860484 542 / 578 The train loss 0.577252494305 543 / 578 The train loss 0.576487560398 544 / 578 The train loss 0.575911939298 545 / 578 The train loss 0.575907390753 546 / 578 The train loss 0.577222573278 547 / 578 The train loss 0.578844621413 548 / 578 The train loss 0.581746198879 549 / 578 The train loss 0.583525388663 550 / 578 The train loss 0.584722765186 551 / 578 The train loss 0.585728283846 552 / 578 The train loss 0.586330879685 553 / 578 The train loss 0.587044821877 554 / 578 The train loss 0.587804353561 555 / 578 The train loss 0.588198184259 556 / 578 The train loss 0.588492301343 557 / 578 The train loss 0.58896061456 558 / 578 The train loss 0.589312715493 559 / 578 The train loss 0.589714253674 560 / 578 The train loss 0.590129439682 561 / 578 The train loss 0.590414571402 562 / 578 The train loss 0.590903209347 563 / 578 The train loss 0.591024620114 564 / 578 The train loss 0.590572965209 565 / 578 The train loss 0.590031266572 566 / 578 The train loss 0.589591580746 567 / 578 The train loss 0.588961298362 568 / 578 The train loss 0.588259306608 569 / 578 The train loss 0.58758009579 570 / 578 The train loss 0.586910627302 571 / 578 The train loss 0.586229875309 572 / 578 The train loss 0.585642747806 573 / 578 The train loss 0.585523730658 574 / 578 The train loss 0.585353416605 575 / 578 The train loss 0.584978169996 576 / 578 The train loss 0.584729053458 577 / 578 The train loss 0.584442895522 578 / 578 The train loss 0.584204631452 Starting epoch 10 Validation: 1 / 30 The valid loss 0.431452602148 2 / 30 The valid loss 0.407194644213 3 / 30 The valid loss 0.382112165292 4 / 30 The valid loss 0.349907979369 5 / 30 The valid loss 0.324079704285 6 / 30 The valid loss 0.356718957424 7 / 30 The valid loss 0.461995278086 8 / 30 The valid loss 0.600109353662 9 / 30 The valid loss 0.742826779683 10 / 30 The valid loss 0.821242010593 11 / 30 The valid loss 0.838978691535 12 / 30 The valid loss 0.850089127819 13 / 30 The valid loss 0.859013580359 14 / 30 The valid loss 0.863023642983 15 / 30 The valid loss 0.846650473277 16 / 30 The valid loss 0.8397260122 17 / 30 The valid loss 0.85358973461 18 / 30 The valid loss 0.86556102501 19 / 30 The valid loss 0.866845943426 20 / 30 The valid loss 0.860858133435 21 / 30 The valid loss 0.846110698723 22 / 30 The valid loss 0.832284716043 23 / 30 The valid loss 0.820254937462 24 / 30 The valid loss 0.811494539181 25 / 30 The valid loss 0.803849895 26 / 30 The valid loss 0.799490036873 27 / 30 The valid loss 0.788260036045 28 / 30 The valid loss 0.769834603582 29 / 30 The valid loss 0.749582783415 30 / 30 The valid loss 0.732721613844 Training 1 / 578 The train loss 0.502926468849 2 / 578 The train loss 0.397271022201 3 / 578 The train loss 0.356914093097 4 / 578 The train loss 0.32880949229 5 / 578 The train loss 0.316450423002 6 / 578 The train loss 0.286282718182 7 / 578 The train loss 0.268071268286 8 / 578 The train loss 0.244527613744 9 / 578 The train loss 0.224851886431 10 / 578 The train loss 0.208919147402 11 / 578 The train loss 0.20004045015 12 / 578 The train loss 0.195041978111 13 / 578 The train loss 0.201460811954 14 / 578 The train loss 0.214101598731 15 / 578 The train loss 0.222971190015 16 / 578 The train loss 0.226673127152 17 / 578 The train loss 0.22700625921 18 / 578 The train loss 0.224129632943 19 / 578 The train loss 0.218493903154 20 / 578 The train loss 0.21108893007 21 / 578 The train loss 0.204109152513 22 / 578 The train loss 0.197117830034 23 / 578 The train loss 0.190525883728 24 / 578 The train loss 0.184018763093 25 / 578 The train loss 0.178632653356 26 / 578 The train loss 0.175000063215 27 / 578 The train loss 0.172434220711 28 / 578 The train loss 0.169450778514 29 / 578 The train loss 0.165361426 30 / 578 The train loss 0.162511385729 31 / 578 The train loss 0.160154001847 32 / 578 The train loss 0.157175661065 33 / 578 The train loss 0.155596541636 34 / 578 The train loss 0.156184404212 35 / 578 The train loss 0.157600733212 36 / 578 The train loss 0.159125955568 37 / 578 The train loss 0.161483832308 38 / 578 The train loss 0.163734167422 39 / 578 The train loss 0.163037609596 40 / 578 The train loss 0.162241910771 41 / 578 The train loss 0.160554532961 42 / 578 The train loss 0.158848800475 43 / 578 The train loss 0.158177263168 44 / 578 The train loss 0.158443953165 45 / 578 The train loss 0.15878684885 46 / 578 The train loss 0.157183402582 47 / 578 The train loss 0.155228628916 48 / 578 The train loss 0.153525311655 49 / 578 The train loss 0.152109107041 50 / 578 The train loss 0.151132398993 51 / 578 The train loss 0.150941237661 52 / 578 The train loss 0.151833658751 53 / 578 The train loss 0.151448771498 54 / 578 The train loss 0.151002481166 55 / 578 The train loss 0.150609674643 56 / 578 The train loss 0.148959505771 57 / 578 The train loss 0.147843381161 58 / 578 The train loss 0.148065080802 59 / 578 The train loss 0.148408591621 60 / 578 The train loss 0.151466856276 61 / 578 The train loss 0.155233644438 62 / 578 The train loss 0.159631443 63 / 578 The train loss 0.162204622158 64 / 578 The train loss 0.163164459518 65 / 578 The train loss 0.163977343417 66 / 578 The train loss 0.162679236947 67 / 578 The train loss 0.161190419277 68 / 578 The train loss 0.159741265699 69 / 578 The train loss 0.158190244944 70 / 578 The train loss 0.157196299253 71 / 578 The train loss 0.156962677574 72 / 578 The train loss 0.158170510653 73 / 578 The train loss 0.159333493356 74 / 578 The train loss 0.160954963265 75 / 578 The train loss 0.164944134305 76 / 578 The train loss 0.16661851162 77 / 578 The train loss 0.165760192633 78 / 578 The train loss 0.165724499724 79 / 578 The train loss 0.167377976536 80 / 578 The train loss 0.168315695086 81 / 578 The train loss 0.169460001413 82 / 578 The train loss 0.173480311772 83 / 578 The train loss 0.177756878491 84 / 578 The train loss 0.181961098109 85 / 578 The train loss 0.185832248234 86 / 578 The train loss 0.189459617964 87 / 578 The train loss 0.194198585319 88 / 578 The train loss 0.197318121664 89 / 578 The train loss 0.204113922636 90 / 578 The train loss 0.208280340706 91 / 578 The train loss 0.207678990909 92 / 578 The train loss 0.207375944511 93 / 578 The train loss 0.206213312484 94 / 578 The train loss 0.205162111868 95 / 578 The train loss 0.204403766009 96 / 578 The train loss 0.203814366483 97 / 578 The train loss 0.203163226817 98 / 578 The train loss 0.203089854965 99 / 578 The train loss 0.203961808385 100 / 578 The train loss 0.204883572571 101 / 578 The train loss 0.2052015262 102 / 578 The train loss 0.20519939894 103 / 578 The train loss 0.204691093487 104 / 578 The train loss 0.20430605941 105 / 578 The train loss 0.20444256007 106 / 578 The train loss 0.204512992101 107 / 578 The train loss 0.203550301875 108 / 578 The train loss 0.203781732289 109 / 578 The train loss 0.203327615634 110 / 578 The train loss 0.202538664131 111 / 578 The train loss 0.202267772745 112 / 578 The train loss 0.202859206158 113 / 578 The train loss 0.203509025375 114 / 578 The train loss 0.205433010042 115 / 578 The train loss 0.208105718604 116 / 578 The train loss 0.213228952814 117 / 578 The train loss 0.217428166165 118 / 578 The train loss 0.22162219279 119 / 578 The train loss 0.225700858292 120 / 578 The train loss 0.231208381585 121 / 578 The train loss 0.236928568922 122 / 578 The train loss 0.24262049553 123 / 578 The train loss 0.250280697202 124 / 578 The train loss 0.259439889492 125 / 578 The train loss 0.269612472564 126 / 578 The train loss 0.279915576919 127 / 578 The train loss 0.290122173932 128 / 578 The train loss 0.29932997658 129 / 578 The train loss 0.311146312935 130 / 578 The train loss 0.323411778332 131 / 578 The train loss 0.332520120955 132 / 578 The train loss 0.335988014513 133 / 578 The train loss 0.337665901176 134 / 578 The train loss 0.338379820586 135 / 578 The train loss 0.340824967209 136 / 578 The train loss 0.344331899795 137 / 578 The train loss 0.348528708664 138 / 578 The train loss 0.350369844559 139 / 578 The train loss 0.35139097544 140 / 578 The train loss 0.351227713084 141 / 578 The train loss 0.350883672661 142 / 578 The train loss 0.351087119614 143 / 578 The train loss 0.352234949 144 / 578 The train loss 0.353750813571 145 / 578 The train loss 0.354764638918 146 / 578 The train loss 0.354640105229 147 / 578 The train loss 0.354883438698 148 / 578 The train loss 0.354937578199 149 / 578 The train loss 0.353706184075 150 / 578 The train loss 0.352771534945 151 / 578 The train loss 0.350688012625 152 / 578 The train loss 0.348852077205 153 / 578 The train loss 0.346933819037 154 / 578 The train loss 0.345171311532 155 / 578 The train loss 0.343986846819 156 / 578 The train loss 0.342919553964 157 / 578 The train loss 0.342225909684 158 / 578 The train loss 0.341663784878 159 / 578 The train loss 0.340813729443 160 / 578 The train loss 0.340112769208 161 / 578 The train loss 0.339186289735 162 / 578 The train loss 0.338059504988 163 / 578 The train loss 0.337339736087 164 / 578 The train loss 0.336338923631 165 / 578 The train loss 0.335318207447 166 / 578 The train loss 0.334132670795 167 / 578 The train loss 0.332891856638 168 / 578 The train loss 0.33737449328 169 / 578 The train loss 0.34457167346 170 / 578 The train loss 0.353339612988 171 / 578 The train loss 0.361321373794 172 / 578 The train loss 0.366411816047 173 / 578 The train loss 0.372278593674 174 / 578 The train loss 0.378986398186 175 / 578 The train loss 0.384696885326 176 / 578 The train loss 0.389919706176 177 / 578 The train loss 0.393289322381 178 / 578 The train loss 0.395944652074 179 / 578 The train loss 0.400035657509 180 / 578 The train loss 0.401981455564 181 / 578 The train loss 0.40299000379 182 / 578 The train loss 0.403030850218 183 / 578 The train loss 0.403292789953 184 / 578 The train loss 0.406463115577 185 / 578 The train loss 0.409465361749 186 / 578 The train loss 0.41242204925 187 / 578 The train loss 0.415421014941 188 / 578 The train loss 0.420019167039 189 / 578 The train loss 0.425887756385 190 / 578 The train loss 0.436833458335 191 / 578 The train loss 0.444697250383 192 / 578 The train loss 0.450983528804 193 / 578 The train loss 0.455687813943 194 / 578 The train loss 0.458538061699 195 / 578 The train loss 0.460153256758 196 / 578 The train loss 0.463434141338 197 / 578 The train loss 0.468821781976 198 / 578 The train loss 0.472984347511 199 / 578 The train loss 0.477633955322 200 / 578 The train loss 0.481487833094 201 / 578 The train loss 0.48752360778 202 / 578 The train loss 0.497093274876 203 / 578 The train loss 0.504288736657 204 / 578 The train loss 0.508079889194 205 / 578 The train loss 0.509320194092 206 / 578 The train loss 0.508752889985 207 / 578 The train loss 0.50798435457 208 / 578 The train loss 0.508530933679 209 / 578 The train loss 0.5117956799 210 / 578 The train loss 0.513345974382 211 / 578 The train loss 0.513428820235 212 / 578 The train loss 0.514205734667 213 / 578 The train loss 0.514450033768 214 / 578 The train loss 0.514017307435 215 / 578 The train loss 0.513105325072 216 / 578 The train loss 0.511641877928 217 / 578 The train loss 0.509928366813 218 / 578 The train loss 0.50807597196 219 / 578 The train loss 0.506623988775 220 / 578 The train loss 0.504789471846 221 / 578 The train loss 0.503180312574 222 / 578 The train loss 0.50183544813 223 / 578 The train loss 0.50060011976 224 / 578 The train loss 0.499912332095 225 / 578 The train loss 0.500087958591 226 / 578 The train loss 0.500384898988 227 / 578 The train loss 0.501538542021 228 / 578 The train loss 0.50263541847 229 / 578 The train loss 0.503778646493 230 / 578 The train loss 0.505048594154 231 / 578 The train loss 0.505629807727 232 / 578 The train loss 0.506651596664 233 / 578 The train loss 0.506034547311 234 / 578 The train loss 0.505981946619 235 / 578 The train loss 0.505121006246 236 / 578 The train loss 0.50402976593 237 / 578 The train loss 0.503233920751 238 / 578 The train loss 0.503311208395 239 / 578 The train loss 0.503954158246 240 / 578 The train loss 0.50334316177 241 / 578 The train loss 0.501736873243 242 / 578 The train loss 0.500831392755 243 / 578 The train loss 0.500497691134 244 / 578 The train loss 0.500601151149 245 / 578 The train loss 0.500739697489 246 / 578 The train loss 0.5011152896 247 / 578 The train loss 0.499926872778 248 / 578 The train loss 0.498503397024 249 / 578 The train loss 0.497947118683 250 / 578 The train loss 0.497652900174 251 / 578 The train loss 0.497599246685 252 / 578 The train loss 0.497870980939 253 / 578 The train loss 0.499564946125 254 / 578 The train loss 0.502187310432 255 / 578 The train loss 0.506170450662 256 / 578 The train loss 0.51012224397 257 / 578 The train loss 0.514707039491 258 / 578 The train loss 0.517516804022 259 / 578 The train loss 0.518733072652 260 / 578 The train loss 0.519616333524 261 / 578 The train loss 0.520465460502 262 / 578 The train loss 0.520198595049 263 / 578 The train loss 0.519798107204 264 / 578 The train loss 0.520600977904 265 / 578 The train loss 0.520801183146 266 / 578 The train loss 0.521115102896 267 / 578 The train loss 0.521261177441 268 / 578 The train loss 0.522563064012 269 / 578 The train loss 0.524542901776 270 / 578 The train loss 0.528266806252 271 / 578 The train loss 0.530757643379 272 / 578 The train loss 0.531821257229 273 / 578 The train loss 0.531990039657 274 / 578 The train loss 0.531726745644 275 / 578 The train loss 0.532084093812 276 / 578 The train loss 0.531535861352 277 / 578 The train loss 0.531057837275 278 / 578 The train loss 0.530600569084 279 / 578 The train loss 0.530455811971 280 / 578 The train loss 0.53016910219 281 / 578 The train loss 0.52967625513 282 / 578 The train loss 0.529090938799 283 / 578 The train loss 0.528565340731 284 / 578 The train loss 0.528427227558 285 / 578 The train loss 0.528579929902 286 / 578 The train loss 0.529406772275 287 / 578 The train loss 0.5299057984 288 / 578 The train loss 0.530439364892 289 / 578 The train loss 0.53152341482 290 / 578 The train loss 0.532547635521 291 / 578 The train loss 0.53378058961 292 / 578 The train loss 0.534653664924 293 / 578 The train loss 0.535394750014 294 / 578 The train loss 0.536525444172 295 / 578 The train loss 0.53719172181 296 / 578 The train loss 0.537580916067 297 / 578 The train loss 0.538761504032 298 / 578 The train loss 0.539771796335 299 / 578 The train loss 0.541044522351 300 / 578 The train loss 0.543194851143 301 / 578 The train loss 0.544695351075 302 / 578 The train loss 0.546143955409 303 / 578 The train loss 0.547238477644 304 / 578 The train loss 0.548705453997 305 / 578 The train loss 0.550204631992 306 / 578 The train loss 0.551563019712 307 / 578 The train loss 0.552912255707 308 / 578 The train loss 0.553503007725 309 / 578 The train loss 0.554480650917 310 / 578 The train loss 0.555680171773 311 / 578 The train loss 0.557187422164 312 / 578 The train loss 0.559527134535 313 / 578 The train loss 0.562161001404 314 / 578 The train loss 0.56471222664 315 / 578 The train loss 0.566724606414 316 / 578 The train loss 0.568327285314 317 / 578 The train loss 0.569758198485 318 / 578 The train loss 0.570973950796 319 / 578 The train loss 0.572016266878 320 / 578 The train loss 0.572258778533 321 / 578 The train loss 0.571791611229 322 / 578 The train loss 0.570486060731 323 / 578 The train loss 0.569382789484 324 / 578 The train loss 0.568194810451 325 / 578 The train loss 0.566808029654 326 / 578 The train loss 0.565431533199 327 / 578 The train loss 0.564028297417 328 / 578 The train loss 0.562663606766 329 / 578 The train loss 0.561273820483 330 / 578 The train loss 0.55999089975 331 / 578 The train loss 0.558717030492 332 / 578 The train loss 0.557221883986 333 / 578 The train loss 0.556630901623 334 / 578 The train loss 0.557192136662 335 / 578 The train loss 0.559427216749 336 / 578 The train loss 0.56127159782 337 / 578 The train loss 0.560149505569 338 / 578 The train loss 0.55893288523 339 / 578 The train loss 0.557664015355 340 / 578 The train loss 0.55636687889 341 / 578 The train loss 0.555195040378 342 / 578 The train loss 0.553764246088 343 / 578 The train loss 0.552600139812 344 / 578 The train loss 0.551737294587 345 / 578 The train loss 0.551125244677 346 / 578 The train loss 0.550238132208 347 / 578 The train loss 0.55008319002 348 / 578 The train loss 0.550210842873 349 / 578 The train loss 0.550897726944 350 / 578 The train loss 0.551483183152 351 / 578 The train loss 0.552481957555 352 / 578 The train loss 0.552857636633 353 / 578 The train loss 0.552929908027 354 / 578 The train loss 0.553176785667 355 / 578 The train loss 0.553756754069 356 / 578 The train loss 0.55430716903 357 / 578 The train loss 0.555019004912 358 / 578 The train loss 0.555897393779 359 / 578 The train loss 0.556692610451 360 / 578 The train loss 0.557082423806 361 / 578 The train loss 0.557468147727 362 / 578 The train loss 0.557749598138 363 / 578 The train loss 0.558109811422 364 / 578 The train loss 0.557733653083 365 / 578 The train loss 0.557012376057 366 / 578 The train loss 0.555853808909 367 / 578 The train loss 0.554615327982 368 / 578 The train loss 0.553451236003 369 / 578 The train loss 0.552414826292 370 / 578 The train loss 0.551205224755 371 / 578 The train loss 0.550298552056 372 / 578 The train loss 0.549445570867 373 / 578 The train loss 0.548459508773 374 / 578 The train loss 0.547470088859 375 / 578 The train loss 0.546559916168 376 / 578 The train loss 0.545626607575 377 / 578 The train loss 0.544763010272 378 / 578 The train loss 0.544045005831 379 / 578 The train loss 0.543316267716 380 / 578 The train loss 0.542172230535 381 / 578 The train loss 0.541384180846 382 / 578 The train loss 0.540897433973 383 / 578 The train loss 0.540318618701 384 / 578 The train loss 0.539338808613 385 / 578 The train loss 0.538176834225 386 / 578 The train loss 0.538045898804 387 / 578 The train loss 0.53832417714 388 / 578 The train loss 0.538982344686 389 / 578 The train loss 0.539827308218 390 / 578 The train loss 0.541113316736 391 / 578 The train loss 0.541850150117 392 / 578 The train loss 0.541976542222 393 / 578 The train loss 0.54215075287 394 / 578 The train loss 0.54174696933 395 / 578 The train loss 0.541296110906 396 / 578 The train loss 0.540830157801 397 / 578 The train loss 0.540517752385 398 / 578 The train loss 0.540585754683 399 / 578 The train loss 0.54096325316 400 / 578 The train loss 0.540837665079 401 / 578 The train loss 0.540626391833 402 / 578 The train loss 0.540422301712 403 / 578 The train loss 0.540482722824 404 / 578 The train loss 0.540592572504 405 / 578 The train loss 0.540955797712 406 / 578 The train loss 0.540878467987 407 / 578 The train loss 0.540531359549 408 / 578 The train loss 0.540049819047 409 / 578 The train loss 0.542116842772 410 / 578 The train loss 0.545813208918 411 / 578 The train loss 0.548178729304 412 / 578 The train loss 0.548825029853 413 / 578 The train loss 0.549640456598 414 / 578 The train loss 0.550888623259 415 / 578 The train loss 0.551832321277 416 / 578 The train loss 0.551594785728 417 / 578 The train loss 0.551101266578 418 / 578 The train loss 0.550504138123 419 / 578 The train loss 0.550003263175 420 / 578 The train loss 0.549519520812 421 / 578 The train loss 0.549327275063 422 / 578 The train loss 0.549809862184 423 / 578 The train loss 0.550717332077 424 / 578 The train loss 0.551164844223 425 / 578 The train loss 0.551551920062 426 / 578 The train loss 0.551563164674 427 / 578 The train loss 0.551230787071 428 / 578 The train loss 0.550761072392 429 / 578 The train loss 0.549971195255 430 / 578 The train loss 0.549137966985 431 / 578 The train loss 0.548193691355 432 / 578 The train loss 0.547138369754 433 / 578 The train loss 0.546102207171 434 / 578 The train loss 0.545260362095 435 / 578 The train loss 0.544359255736 436 / 578 The train loss 0.543400546801 437 / 578 The train loss 0.542456399121 438 / 578 The train loss 0.541933108419 439 / 578 The train loss 0.541559370485 440 / 578 The train loss 0.541074640786 441 / 578 The train loss 0.540541157709 442 / 578 The train loss 0.540104505779 443 / 578 The train loss 0.539569278746 444 / 578 The train loss 0.539723907513 445 / 578 The train loss 0.540068172747 446 / 578 The train loss 0.540939951967 447 / 578 The train loss 0.541373352218 448 / 578 The train loss 0.541576781052 449 / 578 The train loss 0.541600034081 450 / 578 The train loss 0.541389010623 451 / 578 The train loss 0.541210410212 452 / 578 The train loss 0.540864200529 453 / 578 The train loss 0.540514272512 454 / 578 The train loss 0.540175042628 455 / 578 The train loss 0.539868688199 456 / 578 The train loss 0.539420832806 457 / 578 The train loss 0.539451562088 458 / 578 The train loss 0.539128645807 459 / 578 The train loss 0.538709272949 460 / 578 The train loss 0.538567074753 461 / 578 The train loss 0.538461573414 462 / 578 The train loss 0.538271150233 463 / 578 The train loss 0.538467517497 464 / 578 The train loss 0.538825789475 465 / 578 The train loss 0.539156216647 466 / 578 The train loss 0.539159210271 467 / 578 The train loss 0.539113898421 468 / 578 The train loss 0.538762851793 469 / 578 The train loss 0.538669542979 470 / 578 The train loss 0.538713209434 471 / 578 The train loss 0.538689026162 472 / 578 The train loss 0.538342648651 473 / 578 The train loss 0.537976658251 474 / 578 The train loss 0.537519159959 475 / 578 The train loss 0.536960888921 476 / 578 The train loss 0.536377025134 477 / 578 The train loss 0.53551864831 478 / 578 The train loss 0.534579202759 479 / 578 The train loss 0.533743942681 480 / 578 The train loss 0.533001254258 481 / 578 The train loss 0.532258131129 482 / 578 The train loss 0.531580660825 483 / 578 The train loss 0.530854398273 484 / 578 The train loss 0.530086219349 485 / 578 The train loss 0.529246060496 486 / 578 The train loss 0.528538819195 487 / 578 The train loss 0.527756422567 488 / 578 The train loss 0.5269709561 489 / 578 The train loss 0.526267429642 490 / 578 The train loss 0.525649474462 491 / 578 The train loss 0.525061504593 492 / 578 The train loss 0.524586905786 493 / 578 The train loss 0.524036159191 494 / 578 The train loss 0.523297204504 495 / 578 The train loss 0.522533953393 496 / 578 The train loss 0.522003358673 497 / 578 The train loss 0.521532136992 498 / 578 The train loss 0.521088383725 499 / 578 The train loss 0.520529435302 500 / 578 The train loss 0.519834869437 501 / 578 The train loss 0.519192485507 502 / 578 The train loss 0.518698003121 503 / 578 The train loss 0.518098937048 504 / 578 The train loss 0.517396760772 505 / 578 The train loss 0.51666920565 506 / 578 The train loss 0.515959873052 507 / 578 The train loss 0.51539201105 508 / 578 The train loss 0.514506979039 509 / 578 The train loss 0.513717975555 510 / 578 The train loss 0.512848859857 511 / 578 The train loss 0.512015080623 512 / 578 The train loss 0.511137865054 513 / 578 The train loss 0.510257224344 514 / 578 The train loss 0.509555749581 515 / 578 The train loss 0.508860021613 516 / 578 The train loss 0.508207265244 517 / 578 The train loss 0.507579687846 518 / 578 The train loss 0.506813400457 519 / 578 The train loss 0.506032628562 520 / 578 The train loss 0.505219183103 521 / 578 The train loss 0.504375906114 522 / 578 The train loss 0.503491042175 523 / 578 The train loss 0.502693698879 524 / 578 The train loss 0.502019413711 525 / 578 The train loss 0.501337773757 526 / 578 The train loss 0.500660081341 527 / 578 The train loss 0.500132852688 528 / 578 The train loss 0.499747547392 529 / 578 The train loss 0.499268505673 530 / 578 The train loss 0.498955042871 531 / 578 The train loss 0.498581644632 532 / 578 The train loss 0.498285131473 533 / 578 The train loss 0.497792048775 534 / 578 The train loss 0.497295393996 535 / 578 The train loss 0.496603113148 536 / 578 The train loss 0.495922133424 537 / 578 The train loss 0.495234826659 538 / 578 The train loss 0.494854309094 539 / 578 The train loss 0.494329307056 540 / 578 The train loss 0.493790154283 541 / 578 The train loss 0.493231332856 542 / 578 The train loss 0.492793505331 543 / 578 The train loss 0.49241142248 544 / 578 The train loss 0.492182661574 545 / 578 The train loss 0.492191670097 546 / 578 The train loss 0.493627028983 547 / 578 The train loss 0.495333879437 548 / 578 The train loss 0.498402329216 549 / 578 The train loss 0.499950520504 550 / 578 The train loss 0.501209352193 551 / 578 The train loss 0.502378371323 552 / 578 The train loss 0.503302936064 553 / 578 The train loss 0.504183734048 554 / 578 The train loss 0.504728442777 555 / 578 The train loss 0.505115703057 556 / 578 The train loss 0.505589017082 557 / 578 The train loss 0.506321179334 558 / 578 The train loss 0.506714540744 559 / 578 The train loss 0.507164674669 560 / 578 The train loss 0.507709235019 561 / 578 The train loss 0.508250781413 562 / 578 The train loss 0.508774807829 563 / 578 The train loss 0.509013764878 564 / 578 The train loss 0.508568392855 565 / 578 The train loss 0.508019633275 566 / 578 The train loss 0.507582737166 567 / 578 The train loss 0.507250335022 568 / 578 The train loss 0.506841806472 569 / 578 The train loss 0.506217292184 570 / 578 The train loss 0.505590019506 571 / 578 The train loss 0.505095965299 572 / 578 The train loss 0.504620324312 573 / 578 The train loss 0.504661155972 574 / 578 The train loss 0.50454876189 575 / 578 The train loss 0.50425451096 576 / 578 The train loss 0.504095386132 577 / 578 The train loss 0.503836901946 578 / 578 The train loss 0.50361374339 Starting epoch 11 Validation: 1 / 30 The valid loss 0.430199474096 2 / 30 The valid loss 0.405474439263 3 / 30 The valid loss 0.378708531459 4 / 30 The valid loss 0.335772275925 5 / 30 The valid loss 0.304791176319 6 / 30 The valid loss 0.333630969127 7 / 30 The valid loss 0.434569214072 8 / 30 The valid loss 0.569503821433 9 / 30 The valid loss 0.709261609448 10 / 30 The valid loss 0.785347598791 11 / 30 The valid loss 0.803475515409 12 / 30 The valid loss 0.81267079711 13 / 30 The valid loss 0.819989062273 14 / 30 The valid loss 0.823305972985 15 / 30 The valid loss 0.80845990181 16 / 30 The valid loss 0.802762564272 17 / 30 The valid loss 0.816301889279 18 / 30 The valid loss 0.826924731334 19 / 30 The valid loss 0.827641345953 20 / 30 The valid loss 0.822344475985 21 / 30 The valid loss 0.807548173836 22 / 30 The valid loss 0.793464452028 23 / 30 The valid loss 0.781309402507 24 / 30 The valid loss 0.772236317396 25 / 30 The valid loss 0.763131222725 26 / 30 The valid loss 0.757988326825 27 / 30 The valid loss 0.746709728682 28 / 30 The valid loss 0.72892548676 29 / 30 The valid loss 0.710386893359 30 / 30 The valid loss 0.695513339341 Validation MSE(val_loss): 14.7503256502 Test MSE(test_loss): 4.32328072875 Training 1 / 578 The train loss 0.512278556824 2 / 578 The train loss 0.349244430661 3 / 578 The train loss 0.288971463839 4 / 578 The train loss 0.295878715813 5 / 578 The train loss 0.298770433664 6 / 578 The train loss 0.28505191952 7 / 578 The train loss 0.265821622951 8 / 578 The train loss 0.241382758133 9 / 578 The train loss 0.220676879088 10 / 578 The train loss 0.210786686093 11 / 578 The train loss 0.200479431586 12 / 578 The train loss 0.19153811348 13 / 578 The train loss 0.188660960358 14 / 578 The train loss 0.200489352324 15 / 578 The train loss 0.209016596774 16 / 578 The train loss 0.212880739477 17 / 578 The train loss 0.215444997391 18 / 578 The train loss 0.21727362896 19 / 578 The train loss 0.215769311707 20 / 578 The train loss 0.208702167124 21 / 578 The train loss 0.201529500563 22 / 578 The train loss 0.193677536381 23 / 578 The train loss 0.185948542117 24 / 578 The train loss 0.180113698977 25 / 578 The train loss 0.17625305891 26 / 578 The train loss 0.173105245599 27 / 578 The train loss 0.169090303006 28 / 578 The train loss 0.165521704991 29 / 578 The train loss 0.160814644512 30 / 578 The train loss 0.157468881023 31 / 578 The train loss 0.155198795961 32 / 578 The train loss 0.154300472524 33 / 578 The train loss 0.154124374473 34 / 578 The train loss 0.154205853722 35 / 578 The train loss 0.154767780059 36 / 578 The train loss 0.155818691943 37 / 578 The train loss 0.156769397543 38 / 578 The train loss 0.158449092831 39 / 578 The train loss 0.158640940077 40 / 578 The train loss 0.157717495086 41 / 578 The train loss 0.156165115762 42 / 578 The train loss 0.154466170862 43 / 578 The train loss 0.155319696812 44 / 578 The train loss 0.15591901291 45 / 578 The train loss 0.154856041612 46 / 578 The train loss 0.153267329032 47 / 578 The train loss 0.152077309827 48 / 578 The train loss 0.150802651925 49 / 578 The train loss 0.149917819252 50 / 578 The train loss 0.149411267228 51 / 578 The train loss 0.149358483981 52 / 578 The train loss 0.149510354556 53 / 578 The train loss 0.148430714946 54 / 578 The train loss 0.147440147241 55 / 578 The train loss 0.147204932198 56 / 578 The train loss 0.146650934758 57 / 578 The train loss 0.146161097086 58 / 578 The train loss 0.145687359835 59 / 578 The train loss 0.145708335311 60 / 578 The train loss 0.147298770305 61 / 578 The train loss 0.148947768921 62 / 578 The train loss 0.152835213279 63 / 578 The train loss 0.157164195139 64 / 578 The train loss 0.158918918954 65 / 578 The train loss 0.159421344589 66 / 578 The train loss 0.157671650956 67 / 578 The train loss 0.156082015346 68 / 578 The train loss 0.154339976417 69 / 578 The train loss 0.152817661645 70 / 578 The train loss 0.151998471442 71 / 578 The train loss 0.151438927645 72 / 578 The train loss 0.150733069905 73 / 578 The train loss 0.14992965888 74 / 578 The train loss 0.151288320763 75 / 578 The train loss 0.154837395623 76 / 578 The train loss 0.155716931619 77 / 578 The train loss 0.155315473652 78 / 578 The train loss 0.155852681336 79 / 578 The train loss 0.157015788947 80 / 578 The train loss 0.158343408047 81 / 578 The train loss 0.160361805977 82 / 578 The train loss 0.162984708983 83 / 578 The train loss 0.166504697425 84 / 578 The train loss 0.170932436074 85 / 578 The train loss 0.175701643242 86 / 578 The train loss 0.179539125131 87 / 578 The train loss 0.183291183613 88 / 578 The train loss 0.187324980253 89 / 578 The train loss 0.194967071473 90 / 578 The train loss 0.201077928622 91 / 578 The train loss 0.20127690475 92 / 578 The train loss 0.201631363137 93 / 578 The train loss 0.201399362836 94 / 578 The train loss 0.200809285659 95 / 578 The train loss 0.20048131533 96 / 578 The train loss 0.200163442777 97 / 578 The train loss 0.200010050338 98 / 578 The train loss 0.200919932382 99 / 578 The train loss 0.201515204207 100 / 578 The train loss 0.203461069595 101 / 578 The train loss 0.205841058681 102 / 578 The train loss 0.206055669936 103 / 578 The train loss 0.205282815578 104 / 578 The train loss 0.204993626371 105 / 578 The train loss 0.205079015078 106 / 578 The train loss 0.204658318223 107 / 578 The train loss 0.203737418211 108 / 578 The train loss 0.202861211142 109 / 578 The train loss 0.203075157991 110 / 578 The train loss 0.202436698753 111 / 578 The train loss 0.202802917761 112 / 578 The train loss 0.20336268182 113 / 578 The train loss 0.204471216842 114 / 578 The train loss 0.206225377998 115 / 578 The train loss 0.208624028949 116 / 578 The train loss 0.212506432263 117 / 578 The train loss 0.216643602723 118 / 578 The train loss 0.223130807833 119 / 578 The train loss 0.227793832583 120 / 578 The train loss 0.232907489237 121 / 578 The train loss 0.237972908569 122 / 578 The train loss 0.243235098824 123 / 578 The train loss 0.250866648915 124 / 578 The train loss 0.260127036126 125 / 578 The train loss 0.269424164101 126 / 578 The train loss 0.279381592403 127 / 578 The train loss 0.28905893808 128 / 578 The train loss 0.298914042578 129 / 578 The train loss 0.309806157727 130 / 578 The train loss 0.322141466963 131 / 578 The train loss 0.330687000821 132 / 578 The train loss 0.334315839859 133 / 578 The train loss 0.335701431365 134 / 578 The train loss 0.336502517601 135 / 578 The train loss 0.338987449184 136 / 578 The train loss 0.342508613009 137 / 578 The train loss 0.346763296542 138 / 578 The train loss 0.348477939589 139 / 578 The train loss 0.350287468568 140 / 578 The train loss 0.350969517271 141 / 578 The train loss 0.350963401496 142 / 578 The train loss 0.351199127878 143 / 578 The train loss 0.352993488768 144 / 578 The train loss 0.354081523126 145 / 578 The train loss 0.3552160247 146 / 578 The train loss 0.355849105444 147 / 578 The train loss 0.355922472353 148 / 578 The train loss 0.355671094315 149 / 578 The train loss 0.35519891061 150 / 578 The train loss 0.355183719633 151 / 578 The train loss 0.353507768326 152 / 578 The train loss 0.351967725793 153 / 578 The train loss 0.350352366412 154 / 578 The train loss 0.34914642518 155 / 578 The train loss 0.348572703944 156 / 578 The train loss 0.347851042409 157 / 578 The train loss 0.347410050772 158 / 578 The train loss 0.347106155185 159 / 578 The train loss 0.346141041485 160 / 578 The train loss 0.34534178566 161 / 578 The train loss 0.344490195133 162 / 578 The train loss 0.343345834368 163 / 578 The train loss 0.342417757147 164 / 578 The train loss 0.341380552673 165 / 578 The train loss 0.340258600529 166 / 578 The train loss 0.338806878674 167 / 578 The train loss 0.337737893217 168 / 578 The train loss 0.342048657382 169 / 578 The train loss 0.349212890138 170 / 578 The train loss 0.358087533658 171 / 578 The train loss 0.366399807834 172 / 578 The train loss 0.371692000726 173 / 578 The train loss 0.378494415271 174 / 578 The train loss 0.385001215045 175 / 578 The train loss 0.391410870158 176 / 578 The train loss 0.396955887311 177 / 578 The train loss 0.400389888476 178 / 578 The train loss 0.402977443091 179 / 578 The train loss 0.40689219704 180 / 578 The train loss 0.410065660822 181 / 578 The train loss 0.411480028275 182 / 578 The train loss 0.411636244414 183 / 578 The train loss 0.412294121011 184 / 578 The train loss 0.415766393634 185 / 578 The train loss 0.418556686299 186 / 578 The train loss 0.424389503285 187 / 578 The train loss 0.429614794396 188 / 578 The train loss 0.435392559159 189 / 578 The train loss 0.441677631475 190 / 578 The train loss 0.453026355519 191 / 578 The train loss 0.462662017109 192 / 578 The train loss 0.469179399884 193 / 578 The train loss 0.473932103741 194 / 578 The train loss 0.477687744088 195 / 578 The train loss 0.479215719408 196 / 578 The train loss 0.481918192896 197 / 578 The train loss 0.486487590977 198 / 578 The train loss 0.488074885126 199 / 578 The train loss 0.490016072193 200 / 578 The train loss 0.491722825123 201 / 578 The train loss 0.495749582059 202 / 578 The train loss 0.501405297483 203 / 578 The train loss 0.504651045518 204 / 578 The train loss 0.505496581934 205 / 578 The train loss 0.504343983994 206 / 578 The train loss 0.502830796911 207 / 578 The train loss 0.501450184346 208 / 578 The train loss 0.500876365829 209 / 578 The train loss 0.500995208656 210 / 578 The train loss 0.500045959749 211 / 578 The train loss 0.498248227356 212 / 578 The train loss 0.496643370079 213 / 578 The train loss 0.495297438548 214 / 578 The train loss 0.494191317662 215 / 578 The train loss 0.492593903831 216 / 578 The train loss 0.490888080379 217 / 578 The train loss 0.490278478858 218 / 578 The train loss 0.488715854525 219 / 578 The train loss 0.487154567478 220 / 578 The train loss 0.485595297331 221 / 578 The train loss 0.484263389283 222 / 578 The train loss 0.482619610615 223 / 578 The train loss 0.481462024745 224 / 578 The train loss 0.48068911924 225 / 578 The train loss 0.479608263928 226 / 578 The train loss 0.478771212577 227 / 578 The train loss 0.478682635992 228 / 578 The train loss 0.479407597881 229 / 578 The train loss 0.48067371875 230 / 578 The train loss 0.480853651747 231 / 578 The train loss 0.480122638431 232 / 578 The train loss 0.479249291853 233 / 578 The train loss 0.478411274367 234 / 578 The train loss 0.478295972674 235 / 578 The train loss 0.478917699632 236 / 578 The train loss 0.478926000414 237 / 578 The train loss 0.47842416943 238 / 578 The train loss 0.477775381729 239 / 578 The train loss 0.47730684875 240 / 578 The train loss 0.47659010846 241 / 578 The train loss 0.475894407203 242 / 578 The train loss 0.475339750821 243 / 578 The train loss 0.474408004717 244 / 578 The train loss 0.473186394559 245 / 578 The train loss 0.472874960448 246 / 578 The train loss 0.472939395649 247 / 578 The train loss 0.471571497853 248 / 578 The train loss 0.470553343597 249 / 578 The train loss 0.469516777871 250 / 578 The train loss 0.469066550039 251 / 578 The train loss 0.469762138659 252 / 578 The train loss 0.470256879711 253 / 578 The train loss 0.471778176541 254 / 578 The train loss 0.473804835047 255 / 578 The train loss 0.476581795331 256 / 578 The train loss 0.480539886536 257 / 578 The train loss 0.484191781142 258 / 578 The train loss 0.486457186405 259 / 578 The train loss 0.487601744874 260 / 578 The train loss 0.488303046294 261 / 578 The train loss 0.489191664611 262 / 578 The train loss 0.489865194317 263 / 578 The train loss 0.489445440237 264 / 578 The train loss 0.490058563897 265 / 578 The train loss 0.491027126693 266 / 578 The train loss 0.492199611416 267 / 578 The train loss 0.492705023488 268 / 578 The train loss 0.49317779294 269 / 578 The train loss 0.495422498908 270 / 578 The train loss 0.499017911949 271 / 578 The train loss 0.50192090992 272 / 578 The train loss 0.502347354962 273 / 578 The train loss 0.502129440472 274 / 578 The train loss 0.502114832748 275 / 578 The train loss 0.501573445926 276 / 578 The train loss 0.501349029894 277 / 578 The train loss 0.501096837808 278 / 578 The train loss 0.50053263388 279 / 578 The train loss 0.499686555857 280 / 578 The train loss 0.499152851976 281 / 578 The train loss 0.498879776857 282 / 578 The train loss 0.498261909344 283 / 578 The train loss 0.497596754959 284 / 578 The train loss 0.496994340943 285 / 578 The train loss 0.496323943158 286 / 578 The train loss 0.496477256099 287 / 578 The train loss 0.496734151669 288 / 578 The train loss 0.49706805934 289 / 578 The train loss 0.497699680851 290 / 578 The train loss 0.498780670884 291 / 578 The train loss 0.500255431525 292 / 578 The train loss 0.501486191213 293 / 578 The train loss 0.50153076732 294 / 578 The train loss 0.501428531058 295 / 578 The train loss 0.501246075932 296 / 578 The train loss 0.500843766189 297 / 578 The train loss 0.500462918489 298 / 578 The train loss 0.501025808456 299 / 578 The train loss 0.501365120713 300 / 578 The train loss 0.502460100371 301 / 578 The train loss 0.503813102909 302 / 578 The train loss 0.505928522426 303 / 578 The train loss 0.507905796139 304 / 578 The train loss 0.509190294616 305 / 578 The train loss 0.509918409018 306 / 578 The train loss 0.510231783498 307 / 578 The train loss 0.510852460091 308 / 578 The train loss 0.511864656093 309 / 578 The train loss 0.513297121423 310 / 578 The train loss 0.513788596967 311 / 578 The train loss 0.515016927902 312 / 578 The train loss 0.517122097045 313 / 578 The train loss 0.519103621064 314 / 578 The train loss 0.521118888353 315 / 578 The train loss 0.52205281969 316 / 578 The train loss 0.522386967682 317 / 578 The train loss 0.523340388163 318 / 578 The train loss 0.524845098286 319 / 578 The train loss 0.525517046376 320 / 578 The train loss 0.52529957529 321 / 578 The train loss 0.52418683485 322 / 578 The train loss 0.522798777801 323 / 578 The train loss 0.52151838844 324 / 578 The train loss 0.520240350908 325 / 578 The train loss 0.519264401513 326 / 578 The train loss 0.517885664964 327 / 578 The train loss 0.516440820308 328 / 578 The train loss 0.515056228362 329 / 578 The train loss 0.51364148163 330 / 578 The train loss 0.512274454257 331 / 578 The train loss 0.51089301025 332 / 578 The train loss 0.509493858604 333 / 578 The train loss 0.508717958181 334 / 578 The train loss 0.50859810108 335 / 578 The train loss 0.509920511756 336 / 578 The train loss 0.510424866114 337 / 578 The train loss 0.51001661078 338 / 578 The train loss 0.509077503566 339 / 578 The train loss 0.50817211319 340 / 578 The train loss 0.506903652535 341 / 578 The train loss 0.505985985835 342 / 578 The train loss 0.504871672809 343 / 578 The train loss 0.503734052675 344 / 578 The train loss 0.50263663162 345 / 578 The train loss 0.501914423677 346 / 578 The train loss 0.500945481407 347 / 578 The train loss 0.500101210206 348 / 578 The train loss 0.499628755526 349 / 578 The train loss 0.499489956026 350 / 578 The train loss 0.499506595449 351 / 578 The train loss 0.49951576233 352 / 578 The train loss 0.499469725643 353 / 578 The train loss 0.499047819995 354 / 578 The train loss 0.498872824197 355 / 578 The train loss 0.499181581609 356 / 578 The train loss 0.499457402736 357 / 578 The train loss 0.499691404105 358 / 578 The train loss 0.499496058769 359 / 578 The train loss 0.499271539484 360 / 578 The train loss 0.499466478716 361 / 578 The train loss 0.499838579781 362 / 578 The train loss 0.500190619333 363 / 578 The train loss 0.500907417556 364 / 578 The train loss 0.501178620173 365 / 578 The train loss 0.500475592988 366 / 578 The train loss 0.500112686179 367 / 578 The train loss 0.499022762253 368 / 578 The train loss 0.497917499449 369 / 578 The train loss 0.497022763435 370 / 578 The train loss 0.496537528263 371 / 578 The train loss 0.496370331231 372 / 578 The train loss 0.496200950411 373 / 578 The train loss 0.495535830725 374 / 578 The train loss 0.495020452523 375 / 578 The train loss 0.494294317658 376 / 578 The train loss 0.493467897509 377 / 578 The train loss 0.492489471928 378 / 578 The train loss 0.491347057494 379 / 578 The train loss 0.4901752453 380 / 578 The train loss 0.489024979459 381 / 578 The train loss 0.488421749837 382 / 578 The train loss 0.487934103845 383 / 578 The train loss 0.487156179683 384 / 578 The train loss 0.486106032501 385 / 578 The train loss 0.484932907153 386 / 578 The train loss 0.484158924028 387 / 578 The train loss 0.483496024025 388 / 578 The train loss 0.482798494451 389 / 578 The train loss 0.482047150496 390 / 578 The train loss 0.481047968466 391 / 578 The train loss 0.48004802093 392 / 578 The train loss 0.479110667515 393 / 578 The train loss 0.478338832917 394 / 578 The train loss 0.477647420184 395 / 578 The train loss 0.476692184084 396 / 578 The train loss 0.476419237823 397 / 578 The train loss 0.476263864885 398 / 578 The train loss 0.476354137058 399 / 578 The train loss 0.476176528848 400 / 578 The train loss 0.475853389571 401 / 578 The train loss 0.475485150277 402 / 578 The train loss 0.474955633723 403 / 578 The train loss 0.474900932668 404 / 578 The train loss 0.475232165192 405 / 578 The train loss 0.475385003174 406 / 578 The train loss 0.474896014099 407 / 578 The train loss 0.474396004646 408 / 578 The train loss 0.473771010612 409 / 578 The train loss 0.475579911816 410 / 578 The train loss 0.478716923037 411 / 578 The train loss 0.480684723075 412 / 578 The train loss 0.481585966654 413 / 578 The train loss 0.482367157138 414 / 578 The train loss 0.483473455772 415 / 578 The train loss 0.484391325851 416 / 578 The train loss 0.484608125868 417 / 578 The train loss 0.484627809309 418 / 578 The train loss 0.484291855723 419 / 578 The train loss 0.484349561544 420 / 578 The train loss 0.484491340954 421 / 578 The train loss 0.484300616726 422 / 578 The train loss 0.484538858461 423 / 578 The train loss 0.485482965162 424 / 578 The train loss 0.486377992505 425 / 578 The train loss 0.486938587423 426 / 578 The train loss 0.487451693887 427 / 578 The train loss 0.487420694955 428 / 578 The train loss 0.486886403058 429 / 578 The train loss 0.486281525661 430 / 578 The train loss 0.485560083835 431 / 578 The train loss 0.484773035803 432 / 578 The train loss 0.4837395258 433 / 578 The train loss 0.482811965027 434 / 578 The train loss 0.482202487054 435 / 578 The train loss 0.481357638568 436 / 578 The train loss 0.480420718512 437 / 578 The train loss 0.479649695958 438 / 578 The train loss 0.479270550057 439 / 578 The train loss 0.479059976888 440 / 578 The train loss 0.478813219862 441 / 578 The train loss 0.478517501106 442 / 578 The train loss 0.478327511371 443 / 578 The train loss 0.478078009964 444 / 578 The train loss 0.478462219754 445 / 578 The train loss 0.47942507131 446 / 578 The train loss 0.480606950164 447 / 578 The train loss 0.481401609474 448 / 578 The train loss 0.481703886475 449 / 578 The train loss 0.481775046967 450 / 578 The train loss 0.481981786999 451 / 578 The train loss 0.482119788992 452 / 578 The train loss 0.482148291176 453 / 578 The train loss 0.482238848128 454 / 578 The train loss 0.482076966401 455 / 578 The train loss 0.481860154257 456 / 578 The train loss 0.481360537225 457 / 578 The train loss 0.481113479532 458 / 578 The train loss 0.480887844813 459 / 578 The train loss 0.480610605009 460 / 578 The train loss 0.480281286569 461 / 578 The train loss 0.48002208499 462 / 578 The train loss 0.479796179812 463 / 578 The train loss 0.480348776 464 / 578 The train loss 0.481160161932 465 / 578 The train loss 0.481714190314 466 / 578 The train loss 0.482210788584 467 / 578 The train loss 0.482448632743 468 / 578 The train loss 0.482477821963 469 / 578 The train loss 0.482601053913 470 / 578 The train loss 0.482653266425 471 / 578 The train loss 0.48259735909 472 / 578 The train loss 0.482440886273 473 / 578 The train loss 0.482091634501 474 / 578 The train loss 0.481687628185 475 / 578 The train loss 0.481094229421 476 / 578 The train loss 0.480317678439 477 / 578 The train loss 0.479524474517 478 / 578 The train loss 0.478719906948 479 / 578 The train loss 0.47806872032 480 / 578 The train loss 0.477591209731 481 / 578 The train loss 0.476995545151 482 / 578 The train loss 0.476511304553 483 / 578 The train loss 0.476000131808 484 / 578 The train loss 0.475291309459 485 / 578 The train loss 0.474781488385 486 / 578 The train loss 0.474217847783 487 / 578 The train loss 0.473616536389 488 / 578 The train loss 0.472924619573 489 / 578 The train loss 0.472332576145 490 / 578 The train loss 0.471751088995 491 / 578 The train loss 0.471295823994 492 / 578 The train loss 0.470887108604 493 / 578 The train loss 0.470432923056 494 / 578 The train loss 0.469835456291 495 / 578 The train loss 0.469265181695 496 / 578 The train loss 0.468853334171 497 / 578 The train loss 0.46853232992 498 / 578 The train loss 0.468209787332 499 / 578 The train loss 0.467982205243 500 / 578 The train loss 0.467476594593 501 / 578 The train loss 0.466834334434 502 / 578 The train loss 0.466186988694 503 / 578 The train loss 0.46574352782 504 / 578 The train loss 0.465374311902 505 / 578 The train loss 0.464933841473 506 / 578 The train loss 0.464339210932 507 / 578 The train loss 0.463848120535 508 / 578 The train loss 0.46324473009 509 / 578 The train loss 0.462571961293 510 / 578 The train loss 0.461846428916 511 / 578 The train loss 0.461233982525 512 / 578 The train loss 0.460627726872 513 / 578 The train loss 0.459885187051 514 / 578 The train loss 0.459261093531 515 / 578 The train loss 0.458697044122 516 / 578 The train loss 0.458182506837 517 / 578 The train loss 0.45774791509 518 / 578 The train loss 0.457147339363 519 / 578 The train loss 0.456483644209 520 / 578 The train loss 0.455881304118 521 / 578 The train loss 0.455259869283 522 / 578 The train loss 0.454473940809 523 / 578 The train loss 0.453735818445 524 / 578 The train loss 0.45327908138 525 / 578 The train loss 0.452938073834 526 / 578 The train loss 0.452388752085 527 / 578 The train loss 0.45198387407 528 / 578 The train loss 0.451468954896 529 / 578 The train loss 0.451217683666 530 / 578 The train loss 0.450824521837 531 / 578 The train loss 0.45030312023 532 / 578 The train loss 0.449939126909 533 / 578 The train loss 0.449505920583 534 / 578 The train loss 0.448980396931 535 / 578 The train loss 0.448433746874 536 / 578 The train loss 0.447830905544 537 / 578 The train loss 0.447267708488 538 / 578 The train loss 0.446757327104 539 / 578 The train loss 0.446330623763 540 / 578 The train loss 0.445896283913 541 / 578 The train loss 0.445327572225 542 / 578 The train loss 0.444819997972 543 / 578 The train loss 0.444272355964 544 / 578 The train loss 0.443866996166 545 / 578 The train loss 0.443873897786 546 / 578 The train loss 0.445102473063 547 / 578 The train loss 0.44659840207 548 / 578 The train loss 0.449726028997 549 / 578 The train loss 0.451652904206 550 / 578 The train loss 0.452931035611 551 / 578 The train loss 0.454276463909 552 / 578 The train loss 0.455273691474 553 / 578 The train loss 0.456240910804 554 / 578 The train loss 0.457273945337 555 / 578 The train loss 0.458048053223 556 / 578 The train loss 0.458607034359 557 / 578 The train loss 0.459561824655 558 / 578 The train loss 0.4605354003 559 / 578 The train loss 0.461394931598 560 / 578 The train loss 0.462338890737 561 / 578 The train loss 0.462883279355 562 / 578 The train loss 0.463303132059 563 / 578 The train loss 0.463310629411 564 / 578 The train loss 0.462827418756 565 / 578 The train loss 0.462178026024 566 / 578 The train loss 0.461602879501 567 / 578 The train loss 0.461018924284 568 / 578 The train loss 0.46039420949 569 / 578 The train loss 0.459842831292 570 / 578 The train loss 0.459317657224 571 / 578 The train loss 0.458700888854 572 / 578 The train loss 0.458221520646 573 / 578 The train loss 0.458251775472 574 / 578 The train loss 0.458140870748 575 / 578 The train loss 0.457773836064 576 / 578 The train loss 0.457569834197 577 / 578 The train loss 0.457373135208 578 / 578 The train loss 0.457058253112 Starting epoch 12 Validation: 1 / 30 The valid loss 0.422050267458 2 / 30 The valid loss 0.399051129818 3 / 30 The valid loss 0.372101227442 4 / 30 The valid loss 0.326366946101 5 / 30 The valid loss 0.294028794765 6 / 30 The valid loss 0.321254620949 7 / 30 The valid loss 0.420462702002 8 / 30 The valid loss 0.553662531078 9 / 30 The valid loss 0.691204938624 10 / 30 The valid loss 0.764984196424 11 / 30 The valid loss 0.78425634991 12 / 30 The valid loss 0.792947351933 13 / 30 The valid loss 0.799838749262 14 / 30 The valid loss 0.803116943155 15 / 30 The valid loss 0.789502040545 16 / 30 The valid loss 0.784871142358 17 / 30 The valid loss 0.798560994513 18 / 30 The valid loss 0.808650804891 19 / 30 The valid loss 0.809107460474 20 / 30 The valid loss 0.804060122371 21 / 30 The valid loss 0.789191224745 22 / 30 The valid loss 0.775211353194 23 / 30 The valid loss 0.763246499974 24 / 30 The valid loss 0.754111302396 25 / 30 The valid loss 0.74459536314 26 / 30 The valid loss 0.739027722524 27 / 30 The valid loss 0.727914554101 28 / 30 The valid loss 0.710551854223 29 / 30 The valid loss 0.692506796327 30 / 30 The valid loss 0.677911366026 Validation MSE(val_loss): 14.377026435 Test MSE(test_loss): 4.01394664516 Training 1 / 578 The train loss 0.397657185793 2 / 578 The train loss 0.250186063349 3 / 578 The train loss 0.206095623473 4 / 578 The train loss 0.209104469046 5 / 578 The train loss 0.204660765827 6 / 578 The train loss 0.193705922614 7 / 578 The train loss 0.182711301105 8 / 578 The train loss 0.170875897631 9 / 578 The train loss 0.157623284807 10 / 578 The train loss 0.144187627733 11 / 578 The train loss 0.134877782654 12 / 578 The train loss 0.128030076623 13 / 578 The train loss 0.129204782156 14 / 578 The train loss 0.13854912562 15 / 578 The train loss 0.146938010057 16 / 578 The train loss 0.154336035252 17 / 578 The train loss 0.157864306779 18 / 578 The train loss 0.162477221754 19 / 578 The train loss 0.162005559394 20 / 578 The train loss 0.158498890325 21 / 578 The train loss 0.152512376152 22 / 578 The train loss 0.147582762451 23 / 578 The train loss 0.143007030144 24 / 578 The train loss 0.13906727902 25 / 578 The train loss 0.136860833913 26 / 578 The train loss 0.134309160165 27 / 578 The train loss 0.132262888468 28 / 578 The train loss 0.129267912624 29 / 578 The train loss 0.125440845073 30 / 578 The train loss 0.122972811138 31 / 578 The train loss 0.119885586442 32 / 578 The train loss 0.117635335773 33 / 578 The train loss 0.11638285897 34 / 578 The train loss 0.116734974086 35 / 578 The train loss 0.116139211825 36 / 578 The train loss 0.118264459901 37 / 578 The train loss 0.119883040319 38 / 578 The train loss 0.122067694209 39 / 578 The train loss 0.122761299595 40 / 578 The train loss 0.121195160784 41 / 578 The train loss 0.119675007353 42 / 578 The train loss 0.117727831095 43 / 578 The train loss 0.117650168147 44 / 578 The train loss 0.118150440383 45 / 578 The train loss 0.118009034544 46 / 578 The train loss 0.116676099032 47 / 578 The train loss 0.115860578307 48 / 578 The train loss 0.115504991496 49 / 578 The train loss 0.115360076102 50 / 578 The train loss 0.115355515257 51 / 578 The train loss 0.115208607605 52 / 578 The train loss 0.1151842543 53 / 578 The train loss 0.114690487607 54 / 578 The train loss 0.11455221688 55 / 578 The train loss 0.114273523133 56 / 578 The train loss 0.113775545931 57 / 578 The train loss 0.112860738904 58 / 578 The train loss 0.112522640169 59 / 578 The train loss 0.113645137177 60 / 578 The train loss 0.11687908452 61 / 578 The train loss 0.120624281161 62 / 578 The train loss 0.124450044586 63 / 578 The train loss 0.127684973831 64 / 578 The train loss 0.129533548199 65 / 578 The train loss 0.129491919336 66 / 578 The train loss 0.128924825709 67 / 578 The train loss 0.127959460076 68 / 578 The train loss 0.126821568291 69 / 578 The train loss 0.126209449239 70 / 578 The train loss 0.126062514686 71 / 578 The train loss 0.126129230301 72 / 578 The train loss 0.126178424547 73 / 578 The train loss 0.126050040687 74 / 578 The train loss 0.127355181637 75 / 578 The train loss 0.129390708158 76 / 578 The train loss 0.130252652213 77 / 578 The train loss 0.131338347814 78 / 578 The train loss 0.132509865392 79 / 578 The train loss 0.135449609 80 / 578 The train loss 0.138424635632 81 / 578 The train loss 0.141250133468 82 / 578 The train loss 0.145493341246 83 / 578 The train loss 0.149586056685 84 / 578 The train loss 0.153985118183 85 / 578 The train loss 0.158326156306 86 / 578 The train loss 0.161421842657 87 / 578 The train loss 0.165248817711 88 / 578 The train loss 0.168052456215 89 / 578 The train loss 0.174546309694 90 / 578 The train loss 0.179980106776 91 / 578 The train loss 0.180601149757 92 / 578 The train loss 0.181851201691 93 / 578 The train loss 0.182198824381 94 / 578 The train loss 0.18217646184 95 / 578 The train loss 0.182708040939 96 / 578 The train loss 0.184215053877 97 / 578 The train loss 0.186066089179 98 / 578 The train loss 0.188045503191 99 / 578 The train loss 0.190191241489 100 / 578 The train loss 0.192782753669 101 / 578 The train loss 0.19525760143 102 / 578 The train loss 0.196172395137 103 / 578 The train loss 0.195933169495 104 / 578 The train loss 0.195740735839 105 / 578 The train loss 0.196341056448 106 / 578 The train loss 0.196275232219 107 / 578 The train loss 0.196119303145 108 / 578 The train loss 0.195528943195 109 / 578 The train loss 0.194648736968 110 / 578 The train loss 0.193533523956 111 / 578 The train loss 0.193040596506 112 / 578 The train loss 0.192623055573 113 / 578 The train loss 0.193258806832 114 / 578 The train loss 0.19411927844 115 / 578 The train loss 0.196600190097 116 / 578 The train loss 0.199427249263 117 / 578 The train loss 0.2024824519 118 / 578 The train loss 0.207938061585 119 / 578 The train loss 0.21345949389 120 / 578 The train loss 0.218254686054 121 / 578 The train loss 0.224240299253 122 / 578 The train loss 0.230137399627 123 / 578 The train loss 0.238950425683 124 / 578 The train loss 0.249119714053 125 / 578 The train loss 0.258220186859 126 / 578 The train loss 0.267695275239 127 / 578 The train loss 0.278000881969 128 / 578 The train loss 0.288388100133 129 / 578 The train loss 0.299789074462 130 / 578 The train loss 0.311497289086 131 / 578 The train loss 0.319620584401 132 / 578 The train loss 0.323461693034 133 / 578 The train loss 0.325334945259 134 / 578 The train loss 0.326652508312 135 / 578 The train loss 0.328821178857 136 / 578 The train loss 0.331738460289 137 / 578 The train loss 0.335975861163 138 / 578 The train loss 0.337868968349 139 / 578 The train loss 0.339020631216 140 / 578 The train loss 0.339233229762 141 / 578 The train loss 0.338859903395 142 / 578 The train loss 0.339057858692 143 / 578 The train loss 0.340589919421 144 / 578 The train loss 0.342798006333 145 / 578 The train loss 0.344355053044 146 / 578 The train loss 0.344589003272 147 / 578 The train loss 0.344346223361 148 / 578 The train loss 0.343679396712 149 / 578 The train loss 0.343177255793 150 / 578 The train loss 0.342343094572 151 / 578 The train loss 0.340412666441 152 / 578 The train loss 0.338922293832 153 / 578 The train loss 0.337563133099 154 / 578 The train loss 0.336310240138 155 / 578 The train loss 0.335607142578 156 / 578 The train loss 0.334850457784 157 / 578 The train loss 0.334079389146 158 / 578 The train loss 0.333330715452 159 / 578 The train loss 0.332362491637 160 / 578 The train loss 0.331823685463 161 / 578 The train loss 0.331012073303 162 / 578 The train loss 0.329800105274 163 / 578 The train loss 0.329230241866 164 / 578 The train loss 0.328230915432 165 / 578 The train loss 0.326877398541 166 / 578 The train loss 0.32567484513 167 / 578 The train loss 0.324942651779 168 / 578 The train loss 0.33052209319 169 / 578 The train loss 0.338363555916 170 / 578 The train loss 0.345968288687 171 / 578 The train loss 0.353814687101 172 / 578 The train loss 0.35893998882 173 / 578 The train loss 0.363720774112 174 / 578 The train loss 0.368365994965 175 / 578 The train loss 0.373862251022 176 / 578 The train loss 0.378935213942 177 / 578 The train loss 0.381698583963 178 / 578 The train loss 0.383483602571 179 / 578 The train loss 0.385788265366 180 / 578 The train loss 0.387046276675 181 / 578 The train loss 0.387897242972 182 / 578 The train loss 0.387539099022 183 / 578 The train loss 0.388323020231 184 / 578 The train loss 0.391546343797 185 / 578 The train loss 0.395872652672 186 / 578 The train loss 0.405160994119 187 / 578 The train loss 0.418302595117 188 / 578 The train loss 0.430552125433 189 / 578 The train loss 0.440456651959 190 / 578 The train loss 0.451816431296 191 / 578 The train loss 0.460925872862 192 / 578 The train loss 0.468361067721 193 / 578 The train loss 0.474342539347 194 / 578 The train loss 0.478525831354 195 / 578 The train loss 0.480282914161 196 / 578 The train loss 0.48219567581 197 / 578 The train loss 0.483954665152 198 / 578 The train loss 0.483482510435 199 / 578 The train loss 0.483097547244 200 / 578 The train loss 0.482218172532 201 / 578 The train loss 0.481379478707 202 / 578 The train loss 0.481968746899 203 / 578 The train loss 0.481187366188 204 / 578 The train loss 0.480385625023 205 / 578 The train loss 0.479018752004 206 / 578 The train loss 0.479390606555 207 / 578 The train loss 0.480249444194 208 / 578 The train loss 0.478887586335 209 / 578 The train loss 0.477973847708 210 / 578 The train loss 0.476657615868 211 / 578 The train loss 0.475515895669 212 / 578 The train loss 0.473779322281 213 / 578 The train loss 0.472385245537 214 / 578 The train loss 0.470803348069 215 / 578 The train loss 0.46996493265 216 / 578 The train loss 0.468559721019 217 / 578 The train loss 0.466984132073 218 / 578 The train loss 0.465636440527 219 / 578 The train loss 0.46419728743 220 / 578 The train loss 0.462515417232 221 / 578 The train loss 0.46173982216 222 / 578 The train loss 0.461699994784 223 / 578 The train loss 0.461160625394 224 / 578 The train loss 0.460402769253 225 / 578 The train loss 0.46004692629 226 / 578 The train loss 0.459907727118 227 / 578 The train loss 0.460032096732 228 / 578 The train loss 0.460208138304 229 / 578 The train loss 0.460034811682 230 / 578 The train loss 0.459324650742 231 / 578 The train loss 0.459384717418 232 / 578 The train loss 0.459032221296 233 / 578 The train loss 0.45830194206 234 / 578 The train loss 0.457408055679 235 / 578 The train loss 0.456730682815 236 / 578 The train loss 0.455586316616 237 / 578 The train loss 0.4554248646 238 / 578 The train loss 0.456957711658 239 / 578 The train loss 0.457409221024 240 / 578 The train loss 0.457052531109 241 / 578 The train loss 0.45550364412 242 / 578 The train loss 0.454803422417 243 / 578 The train loss 0.455276451192 244 / 578 The train loss 0.455059809229 245 / 578 The train loss 0.455169434891 246 / 578 The train loss 0.454847034839 247 / 578 The train loss 0.453915338481 248 / 578 The train loss 0.452887040368 249 / 578 The train loss 0.452039931807 250 / 578 The train loss 0.452222208515 251 / 578 The train loss 0.45279507952 252 / 578 The train loss 0.453410668993 253 / 578 The train loss 0.454645771769 254 / 578 The train loss 0.457574710617 255 / 578 The train loss 0.460617306668 256 / 578 The train loss 0.463721896071 257 / 578 The train loss 0.46665613059 258 / 578 The train loss 0.468967856377 259 / 578 The train loss 0.469419348669 260 / 578 The train loss 0.470158731493 261 / 578 The train loss 0.470120902277 262 / 578 The train loss 0.47132879861 263 / 578 The train loss 0.471241575244 264 / 578 The train loss 0.471592862512 265 / 578 The train loss 0.472229275178 266 / 578 The train loss 0.473323145545 267 / 578 The train loss 0.474128300122 268 / 578 The train loss 0.474863348181 269 / 578 The train loss 0.476245870835 270 / 578 The train loss 0.478979278949 271 / 578 The train loss 0.481004422455 272 / 578 The train loss 0.48114888722 273 / 578 The train loss 0.480596193587 274 / 578 The train loss 0.479538485748 275 / 578 The train loss 0.478534238948 276 / 578 The train loss 0.477611620533 277 / 578 The train loss 0.47699574837 278 / 578 The train loss 0.476357903908 279 / 578 The train loss 0.47523448866 280 / 578 The train loss 0.474494992316 281 / 578 The train loss 0.473786837385 282 / 578 The train loss 0.473134444234 283 / 578 The train loss 0.473303204409 284 / 578 The train loss 0.473156736783 285 / 578 The train loss 0.473196998513 286 / 578 The train loss 0.473224675231 287 / 578 The train loss 0.472985782807 288 / 578 The train loss 0.473458697722 289 / 578 The train loss 0.473893286151 290 / 578 The train loss 0.474229873164 291 / 578 The train loss 0.474428613072 292 / 578 The train loss 0.475084784823 293 / 578 The train loss 0.475118619603 294 / 578 The train loss 0.475088335333 295 / 578 The train loss 0.474496528367 296 / 578 The train loss 0.474013005169 297 / 578 The train loss 0.473340784773 298 / 578 The train loss 0.472842870455 299 / 578 The train loss 0.473371653565 300 / 578 The train loss 0.473871631734 301 / 578 The train loss 0.474628258299 302 / 578 The train loss 0.475321510486 303 / 578 The train loss 0.475305192062 304 / 578 The train loss 0.475473237187 305 / 578 The train loss 0.476230671477 306 / 578 The train loss 0.477563414224 307 / 578 The train loss 0.478299487406 308 / 578 The train loss 0.480062443823 309 / 578 The train loss 0.481573297284 310 / 578 The train loss 0.481781575288 311 / 578 The train loss 0.482264273144 312 / 578 The train loss 0.482008210945 313 / 578 The train loss 0.481264271211 314 / 578 The train loss 0.480672176583 315 / 578 The train loss 0.480420806874 316 / 578 The train loss 0.48090508902 317 / 578 The train loss 0.482146935708 318 / 578 The train loss 0.482289752882 319 / 578 The train loss 0.482303041474 320 / 578 The train loss 0.481912113365 321 / 578 The train loss 0.481000587634 322 / 578 The train loss 0.479659508162 323 / 578 The train loss 0.478259428494 324 / 578 The train loss 0.477058423189 325 / 578 The train loss 0.476517377301 326 / 578 The train loss 0.476205802345 327 / 578 The train loss 0.475471286359 328 / 578 The train loss 0.47441676635 329 / 578 The train loss 0.473435657037 330 / 578 The train loss 0.472203625795 331 / 578 The train loss 0.470980850535 332 / 578 The train loss 0.469838154816 333 / 578 The train loss 0.46901890433 334 / 578 The train loss 0.468666970942 335 / 578 The train loss 0.468694071305 336 / 578 The train loss 0.467599445632 337 / 578 The train loss 0.466349680308 338 / 578 The train loss 0.46534567215 339 / 578 The train loss 0.464407113878 340 / 578 The train loss 0.463198259396 341 / 578 The train loss 0.462186669117 342 / 578 The train loss 0.461026229367 343 / 578 The train loss 0.460073541329 344 / 578 The train loss 0.459201383004 345 / 578 The train loss 0.458358075578 346 / 578 The train loss 0.457485844817 347 / 578 The train loss 0.457026395753 348 / 578 The train loss 0.456823345413 349 / 578 The train loss 0.456593612641 350 / 578 The train loss 0.456088381165 351 / 578 The train loss 0.45565253251 352 / 578 The train loss 0.455263823705 353 / 578 The train loss 0.454946925901 354 / 578 The train loss 0.454875024962 355 / 578 The train loss 0.454959444271 356 / 578 The train loss 0.45503313581 357 / 578 The train loss 0.454741919251 358 / 578 The train loss 0.454595840853 359 / 578 The train loss 0.454245167533 360 / 578 The train loss 0.4540631498 361 / 578 The train loss 0.453616133285 362 / 578 The train loss 0.45331053972 363 / 578 The train loss 0.453011410798 364 / 578 The train loss 0.452029048649 365 / 578 The train loss 0.45094328538 366 / 578 The train loss 0.450031384001 367 / 578 The train loss 0.448869141083 368 / 578 The train loss 0.447832617398 369 / 578 The train loss 0.447112890011 370 / 578 The train loss 0.446421442216 371 / 578 The train loss 0.445722323117 372 / 578 The train loss 0.445321387957 373 / 578 The train loss 0.444798916953 374 / 578 The train loss 0.444393555086 375 / 578 The train loss 0.444221566573 376 / 578 The train loss 0.444010859212 377 / 578 The train loss 0.443514047685 378 / 578 The train loss 0.44298370527 379 / 578 The train loss 0.442213129611 380 / 578 The train loss 0.441355346315 381 / 578 The train loss 0.440900962668 382 / 578 The train loss 0.440906015771 383 / 578 The train loss 0.440529060985 384 / 578 The train loss 0.439745335427 385 / 578 The train loss 0.438849309082 386 / 578 The train loss 0.438270663539 387 / 578 The train loss 0.437934947756 388 / 578 The train loss 0.437735361186 389 / 578 The train loss 0.437492525856 390 / 578 The train loss 0.436909318233 391 / 578 The train loss 0.436363263084 392 / 578 The train loss 0.435587002834 393 / 578 The train loss 0.434804102728 394 / 578 The train loss 0.433995571077 395 / 578 The train loss 0.433112037554 396 / 578 The train loss 0.432674300335 397 / 578 The train loss 0.432339373999 398 / 578 The train loss 0.432129187495 399 / 578 The train loss 0.431829637598 400 / 578 The train loss 0.431400249447 401 / 578 The train loss 0.430862991139 402 / 578 The train loss 0.430253378041 403 / 578 The train loss 0.429861095935 404 / 578 The train loss 0.429755312524 405 / 578 The train loss 0.42963350481 406 / 578 The train loss 0.429267417984 407 / 578 The train loss 0.428800047611 408 / 578 The train loss 0.42817552106 409 / 578 The train loss 0.429657600724 410 / 578 The train loss 0.432446134776 411 / 578 The train loss 0.434231899765 412 / 578 The train loss 0.434827414635 413 / 578 The train loss 0.435230204141 414 / 578 The train loss 0.436170608629 415 / 578 The train loss 0.436991723003 416 / 578 The train loss 0.437020347843 417 / 578 The train loss 0.436929962363 418 / 578 The train loss 0.436436984915 419 / 578 The train loss 0.435914646394 420 / 578 The train loss 0.435834465916 421 / 578 The train loss 0.435789407569 422 / 578 The train loss 0.435736507244 423 / 578 The train loss 0.436203213796 424 / 578 The train loss 0.436526531384 425 / 578 The train loss 0.436674539548 426 / 578 The train loss 0.436408019047 427 / 578 The train loss 0.43584154996 428 / 578 The train loss 0.435149345682 429 / 578 The train loss 0.434476199853 430 / 578 The train loss 0.433593714961 431 / 578 The train loss 0.432702793526 432 / 578 The train loss 0.431754028226 433 / 578 The train loss 0.4308638155 434 / 578 The train loss 0.430273747363 435 / 578 The train loss 0.429751389097 436 / 578 The train loss 0.429327022635 437 / 578 The train loss 0.428792316432 438 / 578 The train loss 0.428682481632 439 / 578 The train loss 0.428587880698 440 / 578 The train loss 0.428369938083 441 / 578 The train loss 0.428010456931 442 / 578 The train loss 0.42751681465 443 / 578 The train loss 0.427083531542 444 / 578 The train loss 0.427165533215 445 / 578 The train loss 0.42736810671 446 / 578 The train loss 0.428025018983 447 / 578 The train loss 0.428497164683 448 / 578 The train loss 0.428676994892 449 / 578 The train loss 0.428600213617 450 / 578 The train loss 0.428649980997 451 / 578 The train loss 0.428821788634 452 / 578 The train loss 0.429158447566 453 / 578 The train loss 0.429244916541 454 / 578 The train loss 0.429076636128 455 / 578 The train loss 0.428828568552 456 / 578 The train loss 0.428398012372 457 / 578 The train loss 0.427931772712 458 / 578 The train loss 0.427399759966 459 / 578 The train loss 0.426878912687 460 / 578 The train loss 0.426245465591 461 / 578 The train loss 0.425577383812 462 / 578 The train loss 0.425041914139 463 / 578 The train loss 0.425259565713 464 / 578 The train loss 0.425571146734 465 / 578 The train loss 0.425266450539 466 / 578 The train loss 0.424627709374 467 / 578 The train loss 0.42390664489 468 / 578 The train loss 0.423281349998 469 / 578 The train loss 0.422914106192 470 / 578 The train loss 0.422525647084 471 / 578 The train loss 0.421973892671 472 / 578 The train loss 0.42144396894 473 / 578 The train loss 0.420982652334 474 / 578 The train loss 0.420460957979 475 / 578 The train loss 0.419969568653 476 / 578 The train loss 0.419332949092 477 / 578 The train loss 0.418817794493 478 / 578 The train loss 0.418110134548 479 / 578 The train loss 0.417553712522 480 / 578 The train loss 0.417335063211 481 / 578 The train loss 0.417126360522 482 / 578 The train loss 0.416839659021 483 / 578 The train loss 0.416522395652 484 / 578 The train loss 0.416212145447 485 / 578 The train loss 0.415759266091 486 / 578 The train loss 0.415269146662 487 / 578 The train loss 0.414777584626 488 / 578 The train loss 0.414258660833 489 / 578 The train loss 0.413671880069 490 / 578 The train loss 0.41312964646 491 / 578 The train loss 0.412696125756 492 / 578 The train loss 0.412312660692 493 / 578 The train loss 0.412149949978 494 / 578 The train loss 0.411836054093 495 / 578 The train loss 0.411546344882 496 / 578 The train loss 0.411611503163 497 / 578 The train loss 0.411789303567 498 / 578 The train loss 0.411962662708 499 / 578 The train loss 0.412059814148 500 / 578 The train loss 0.411967699759 501 / 578 The train loss 0.411653221419 502 / 578 The train loss 0.411352367864 503 / 578 The train loss 0.411134704555 504 / 578 The train loss 0.411037524652 505 / 578 The train loss 0.410706998032 506 / 578 The train loss 0.41029244263 507 / 578 The train loss 0.409947903632 508 / 578 The train loss 0.409541352843 509 / 578 The train loss 0.408921810485 510 / 578 The train loss 0.408337540903 511 / 578 The train loss 0.407788895238 512 / 578 The train loss 0.407201651156 513 / 578 The train loss 0.406692936835 514 / 578 The train loss 0.406108912856 515 / 578 The train loss 0.405511122805 516 / 578 The train loss 0.405053233894 517 / 578 The train loss 0.404803438714 518 / 578 The train loss 0.404367945693 519 / 578 The train loss 0.403886721981 520 / 578 The train loss 0.403484141175 521 / 578 The train loss 0.402910339033 522 / 578 The train loss 0.402226862937 523 / 578 The train loss 0.401642845619 524 / 578 The train loss 0.40118029398 525 / 578 The train loss 0.400801593718 526 / 578 The train loss 0.400390732679 527 / 578 The train loss 0.400008525127 528 / 578 The train loss 0.399665656923 529 / 578 The train loss 0.399375945724 530 / 578 The train loss 0.399220336327 531 / 578 The train loss 0.398928100303 532 / 578 The train loss 0.39862735239 533 / 578 The train loss 0.398332733766 534 / 578 The train loss 0.397913179371 535 / 578 The train loss 0.397354168842 536 / 578 The train loss 0.396791152656 537 / 578 The train loss 0.396276591918 538 / 578 The train loss 0.39588007288 539 / 578 The train loss 0.395498810277 540 / 578 The train loss 0.395075033895 541 / 578 The train loss 0.39475003371 542 / 578 The train loss 0.394496576331 543 / 578 The train loss 0.394051312865 544 / 578 The train loss 0.393769727924 545 / 578 The train loss 0.393886181644 546 / 578 The train loss 0.395013360545 547 / 578 The train loss 0.396783141607 548 / 578 The train loss 0.40000007515 549 / 578 The train loss 0.402007461693 550 / 578 The train loss 0.40328235996 551 / 578 The train loss 0.404372443535 552 / 578 The train loss 0.40501066709 553 / 578 The train loss 0.40573473338 554 / 578 The train loss 0.406523619869 555 / 578 The train loss 0.407152899872 556 / 578 The train loss 0.407509473494 557 / 578 The train loss 0.407839155296 558 / 578 The train loss 0.408442205439 559 / 578 The train loss 0.408822887925 560 / 578 The train loss 0.408912352965 561 / 578 The train loss 0.408935909085 562 / 578 The train loss 0.408954013086 563 / 578 The train loss 0.40879147611 564 / 578 The train loss 0.408322072449 565 / 578 The train loss 0.407909708305 566 / 578 The train loss 0.407542073882 567 / 578 The train loss 0.407281169382 568 / 578 The train loss 0.406905572481 569 / 578 The train loss 0.406452333506 570 / 578 The train loss 0.40592790595 571 / 578 The train loss 0.405445362466 572 / 578 The train loss 0.404979709705 573 / 578 The train loss 0.404891237838 574 / 578 The train loss 0.404667094285 575 / 578 The train loss 0.404456266877 576 / 578 The train loss 0.404354415459 577 / 578 The train loss 0.404115185619 578 / 578 The train loss 0.403742483421 Starting epoch 13 Validation: 1 / 30 The valid loss 0.410781592131 2 / 30 The valid loss 0.380826771259 3 / 30 The valid loss 0.347154537837 4 / 30 The valid loss 0.29259512201 5 / 30 The valid loss 0.259649071097 6 / 30 The valid loss 0.283547294637 7 / 30 The valid loss 0.378085002303 8 / 30 The valid loss 0.509076507762 9 / 30 The valid loss 0.644111211101 10 / 30 The valid loss 0.716082812846 11 / 30 The valid loss 0.744709037922 12 / 30 The valid loss 0.756708384802 13 / 30 The valid loss 0.764705631595 14 / 30 The valid loss 0.768859655729 15 / 30 The valid loss 0.757852213581 16 / 30 The valid loss 0.754644748755 17 / 30 The valid loss 0.767736765392 18 / 30 The valid loss 0.777108951575 19 / 30 The valid loss 0.777215839217 20 / 30 The valid loss 0.773005921394 21 / 30 The valid loss 0.757276513037 22 / 30 The valid loss 0.74174421687 23 / 30 The valid loss 0.728312975039 24 / 30 The valid loss 0.717687033738 25 / 30 The valid loss 0.705975443721 26 / 30 The valid loss 0.69834116044 27 / 30 The valid loss 0.686430512203 28 / 30 The valid loss 0.66986360667 29 / 30 The valid loss 0.653493264112 30 / 30 The valid loss 0.640425000091 Validation MSE(val_loss): 13.5820222113 Test MSE(test_loss): 3.57910447677 Training 1 / 578 The train loss 0.368470460176 2 / 578 The train loss 0.240226142108 3 / 578 The train loss 0.202078978221 4 / 578 The train loss 0.21113075316 5 / 578 The train loss 0.213841882348 6 / 578 The train loss 0.20675902317 7 / 578 The train loss 0.193302235433 8 / 578 The train loss 0.174564985093 9 / 578 The train loss 0.161717348215 10 / 578 The train loss 0.149543242529 11 / 578 The train loss 0.140501940792 12 / 578 The train loss 0.132322738568 13 / 578 The train loss 0.131736484285 14 / 578 The train loss 0.147245638605 15 / 578 The train loss 0.157572737833 16 / 578 The train loss 0.167349963915 17 / 578 The train loss 0.169911931105 18 / 578 The train loss 0.173575416621 19 / 578 The train loss 0.172687494833 20 / 578 The train loss 0.166697440296 21 / 578 The train loss 0.161446570463 22 / 578 The train loss 0.155868079683 23 / 578 The train loss 0.153721928111 24 / 578 The train loss 0.152342348825 25 / 578 The train loss 0.150446001142 26 / 578 The train loss 0.149909641164 27 / 578 The train loss 0.14837535181 28 / 578 The train loss 0.144844022979 29 / 578 The train loss 0.141039618387 30 / 578 The train loss 0.139950921386 31 / 578 The train loss 0.137995934534 32 / 578 The train loss 0.135384467547 33 / 578 The train loss 0.134979814404 34 / 578 The train loss 0.133930447895 35 / 578 The train loss 0.132873545481 36 / 578 The train loss 0.132060709409 37 / 578 The train loss 0.131840285317 38 / 578 The train loss 0.131493933009 39 / 578 The train loss 0.129769840397 40 / 578 The train loss 0.127396366466 41 / 578 The train loss 0.124876934263 42 / 578 The train loss 0.122598890526 43 / 578 The train loss 0.12145129379 44 / 578 The train loss 0.12096267274 45 / 578 The train loss 0.120125612575 46 / 578 The train loss 0.118844224343 47 / 578 The train loss 0.117897455164 48 / 578 The train loss 0.118260172312 49 / 578 The train loss 0.11728263749 50 / 578 The train loss 0.117441848777 51 / 578 The train loss 0.117484884337 52 / 578 The train loss 0.11739296004 53 / 578 The train loss 0.116555299645 54 / 578 The train loss 0.115896022396 55 / 578 The train loss 0.114726778831 56 / 578 The train loss 0.113228443578 57 / 578 The train loss 0.112218254785 58 / 578 The train loss 0.111649209601 59 / 578 The train loss 0.112689692117 60 / 578 The train loss 0.116480102173 61 / 578 The train loss 0.120579980955 62 / 578 The train loss 0.123838492489 63 / 578 The train loss 0.126681078551 64 / 578 The train loss 0.127825316304 65 / 578 The train loss 0.127323560216 66 / 578 The train loss 0.126646268125 67 / 578 The train loss 0.12628945522 68 / 578 The train loss 0.126057890655 69 / 578 The train loss 0.125139804449 70 / 578 The train loss 0.123836037756 71 / 578 The train loss 0.123026938007 72 / 578 The train loss 0.12217493689 73 / 578 The train loss 0.121592153961 74 / 578 The train loss 0.122674838438 75 / 578 The train loss 0.125733399714 76 / 578 The train loss 0.127193206358 77 / 578 The train loss 0.128002621258 78 / 578 The train loss 0.12897403703 79 / 578 The train loss 0.132316755838 80 / 578 The train loss 0.136118912813 81 / 578 The train loss 0.140677995267 82 / 578 The train loss 0.147249091803 83 / 578 The train loss 0.152919557551 84 / 578 The train loss 0.158267390981 85 / 578 The train loss 0.161326116784 86 / 578 The train loss 0.163272715693 87 / 578 The train loss 0.166061095572 88 / 578 The train loss 0.169324901738 89 / 578 The train loss 0.173951494999 90 / 578 The train loss 0.17691302641 91 / 578 The train loss 0.176551133005 92 / 578 The train loss 0.17744354023 93 / 578 The train loss 0.176924641675 94 / 578 The train loss 0.176600078577 95 / 578 The train loss 0.176674214888 96 / 578 The train loss 0.177128637772 97 / 578 The train loss 0.179626753285 98 / 578 The train loss 0.181743714217 99 / 578 The train loss 0.183342154825 100 / 578 The train loss 0.185201155785 101 / 578 The train loss 0.186544646076 102 / 578 The train loss 0.187485424564 103 / 578 The train loss 0.188029513143 104 / 578 The train loss 0.188784627333 105 / 578 The train loss 0.189977310332 106 / 578 The train loss 0.191112288274 107 / 578 The train loss 0.191412267232 108 / 578 The train loss 0.191574773389 109 / 578 The train loss 0.19090749537 110 / 578 The train loss 0.189652082984 111 / 578 The train loss 0.188541388643 112 / 578 The train loss 0.187496015918 113 / 578 The train loss 0.186834448784 114 / 578 The train loss 0.186732810366 115 / 578 The train loss 0.187976458329 116 / 578 The train loss 0.190144079047 117 / 578 The train loss 0.192475423392 118 / 578 The train loss 0.196370578971 119 / 578 The train loss 0.201201737475 120 / 578 The train loss 0.205335644927 121 / 578 The train loss 0.210228489496 122 / 578 The train loss 0.215577308318 123 / 578 The train loss 0.223523606175 124 / 578 The train loss 0.235018929211 125 / 578 The train loss 0.245604332015 126 / 578 The train loss 0.254483878657 127 / 578 The train loss 0.263298129516 128 / 578 The train loss 0.274028199769 129 / 578 The train loss 0.285878331206 130 / 578 The train loss 0.297982630406 131 / 578 The train loss 0.307149654169 132 / 578 The train loss 0.310872359752 133 / 578 The train loss 0.312472043671 134 / 578 The train loss 0.314189815296 135 / 578 The train loss 0.315929490148 136 / 578 The train loss 0.318293765667 137 / 578 The train loss 0.321717971909 138 / 578 The train loss 0.32308597802 139 / 578 The train loss 0.323994732629 140 / 578 The train loss 0.324328554142 141 / 578 The train loss 0.324157901095 142 / 578 The train loss 0.32417263955 143 / 578 The train loss 0.325466491582 144 / 578 The train loss 0.325885422774 145 / 578 The train loss 0.326041490906 146 / 578 The train loss 0.32650279329 147 / 578 The train loss 0.326135412403 148 / 578 The train loss 0.325233964841 149 / 578 The train loss 0.324434713867 150 / 578 The train loss 0.323398213126 151 / 578 The train loss 0.321818727075 152 / 578 The train loss 0.320205955852 153 / 578 The train loss 0.318724684308 154 / 578 The train loss 0.317528051252 155 / 578 The train loss 0.316450177121 156 / 578 The train loss 0.315533530672 157 / 578 The train loss 0.314584240068 158 / 578 The train loss 0.313894912164 159 / 578 The train loss 0.312973322675 160 / 578 The train loss 0.311917337531 161 / 578 The train loss 0.310680333444 162 / 578 The train loss 0.309418608274 163 / 578 The train loss 0.30923833866 164 / 578 The train loss 0.308071093577 165 / 578 The train loss 0.306906773906 166 / 578 The train loss 0.30594904594 167 / 578 The train loss 0.305117566507 168 / 578 The train loss 0.308374631341 169 / 578 The train loss 0.313147214115 170 / 578 The train loss 0.318272790565 171 / 578 The train loss 0.322251025075 172 / 578 The train loss 0.325355015598 173 / 578 The train loss 0.329080555211 174 / 578 The train loss 0.332454708533 175 / 578 The train loss 0.336157878201 176 / 578 The train loss 0.340075199312 177 / 578 The train loss 0.341731716758 178 / 578 The train loss 0.343620049223 179 / 578 The train loss 0.345874241017 180 / 578 The train loss 0.347423365785 181 / 578 The train loss 0.347832251392 182 / 578 The train loss 0.347367215687 183 / 578 The train loss 0.346868910659 184 / 578 The train loss 0.349202405084 185 / 578 The train loss 0.351698558627 186 / 578 The train loss 0.353659897953 187 / 578 The train loss 0.356766317528 188 / 578 The train loss 0.363754112581 189 / 578 The train loss 0.371446470646 190 / 578 The train loss 0.380208492191 191 / 578 The train loss 0.386867392038 192 / 578 The train loss 0.391591900251 193 / 578 The train loss 0.395701319356 194 / 578 The train loss 0.398569503309 195 / 578 The train loss 0.399230638137 196 / 578 The train loss 0.39890955532 197 / 578 The train loss 0.398208997491 198 / 578 The train loss 0.398340224232 199 / 578 The train loss 0.401020193435 200 / 578 The train loss 0.402352152681 201 / 578 The train loss 0.402779023066 202 / 578 The train loss 0.405062416615 203 / 578 The train loss 0.405900450974 204 / 578 The train loss 0.406086898235 205 / 578 The train loss 0.405033683404 206 / 578 The train loss 0.404556144996 207 / 578 The train loss 0.404106909013 208 / 578 The train loss 0.403325649972 209 / 578 The train loss 0.402974727327 210 / 578 The train loss 0.401874604529 211 / 578 The train loss 0.400387943347 212 / 578 The train loss 0.399208349712 213 / 578 The train loss 0.398165355327 214 / 578 The train loss 0.396858804267 215 / 578 The train loss 0.396071393814 216 / 578 The train loss 0.395621057591 217 / 578 The train loss 0.395238186074 218 / 578 The train loss 0.394288208566 219 / 578 The train loss 0.393919802071 220 / 578 The train loss 0.393051222839 221 / 578 The train loss 0.393026731731 222 / 578 The train loss 0.393711920584 223 / 578 The train loss 0.394996484625 224 / 578 The train loss 0.394944845624 225 / 578 The train loss 0.395313961829 226 / 578 The train loss 0.396802184123 227 / 578 The train loss 0.398343676307 228 / 578 The train loss 0.39905770181 229 / 578 The train loss 0.399930203643 230 / 578 The train loss 0.400705321255 231 / 578 The train loss 0.40182058693 232 / 578 The train loss 0.403006141612 233 / 578 The train loss 0.403982794136 234 / 578 The train loss 0.404533930115 235 / 578 The train loss 0.404869738213 236 / 578 The train loss 0.404525115257 237 / 578 The train loss 0.405557755357 238 / 578 The train loss 0.409781230277 239 / 578 The train loss 0.413045765146 240 / 578 The train loss 0.415918808361 241 / 578 The train loss 0.416433590259 242 / 578 The train loss 0.415795637487 243 / 578 The train loss 0.415616537249 244 / 578 The train loss 0.415240614438 245 / 578 The train loss 0.415876147022 246 / 578 The train loss 0.416836436013 247 / 578 The train loss 0.416183716184 248 / 578 The train loss 0.415374498867 249 / 578 The train loss 0.414583109162 250 / 578 The train loss 0.414727124028 251 / 578 The train loss 0.414973489369 252 / 578 The train loss 0.415232674285 253 / 578 The train loss 0.416311037686 254 / 578 The train loss 0.419333450254 255 / 578 The train loss 0.421887964178 256 / 578 The train loss 0.426063996558 257 / 578 The train loss 0.429056847888 258 / 578 The train loss 0.431172465913 259 / 578 The train loss 0.431362756251 260 / 578 The train loss 0.432385800714 261 / 578 The train loss 0.433228900884 262 / 578 The train loss 0.433361540364 263 / 578 The train loss 0.433153050077 264 / 578 The train loss 0.433565858236 265 / 578 The train loss 0.434857539235 266 / 578 The train loss 0.435388501026 267 / 578 The train loss 0.435697750478 268 / 578 The train loss 0.436227556089 269 / 578 The train loss 0.43727289062 270 / 578 The train loss 0.440015284048 271 / 578 The train loss 0.440947430208 272 / 578 The train loss 0.44085622496 273 / 578 The train loss 0.439961406167 274 / 578 The train loss 0.439337638764 275 / 578 The train loss 0.438378827782 276 / 578 The train loss 0.437600249067 277 / 578 The train loss 0.436404475929 278 / 578 The train loss 0.43532189753 279 / 578 The train loss 0.434219242364 280 / 578 The train loss 0.433503973318 281 / 578 The train loss 0.433350689305 282 / 578 The train loss 0.432272228148 283 / 578 The train loss 0.431929219205 284 / 578 The train loss 0.431509053512 285 / 578 The train loss 0.430863624491 286 / 578 The train loss 0.430173263729 287 / 578 The train loss 0.429731763244 288 / 578 The train loss 0.429501543384 289 / 578 The train loss 0.429304465748 290 / 578 The train loss 0.429307364743 291 / 578 The train loss 0.42937213996 292 / 578 The train loss 0.429609916026 293 / 578 The train loss 0.429268981861 294 / 578 The train loss 0.42876240575 295 / 578 The train loss 0.42810884865 296 / 578 The train loss 0.427532517823 297 / 578 The train loss 0.426723695257 298 / 578 The train loss 0.425927220269 299 / 578 The train loss 0.425302847267 300 / 578 The train loss 0.42555860032 301 / 578 The train loss 0.425626839112 302 / 578 The train loss 0.42595609568 303 / 578 The train loss 0.426642945885 304 / 578 The train loss 0.42709998727 305 / 578 The train loss 0.427611013703 306 / 578 The train loss 0.428163291381 307 / 578 The train loss 0.427917839644 308 / 578 The train loss 0.427721996774 309 / 578 The train loss 0.427534290785 310 / 578 The train loss 0.426957224496 311 / 578 The train loss 0.42623634171 312 / 578 The train loss 0.425504524637 313 / 578 The train loss 0.424697590516 314 / 578 The train loss 0.423768960456 315 / 578 The train loss 0.42333622631 316 / 578 The train loss 0.424189880579 317 / 578 The train loss 0.425672050618 318 / 578 The train loss 0.425615166337 319 / 578 The train loss 0.425294019486 320 / 578 The train loss 0.424604065897 321 / 578 The train loss 0.423450957467 322 / 578 The train loss 0.422378498211 323 / 578 The train loss 0.421583413668 324 / 578 The train loss 0.421298548263 325 / 578 The train loss 0.421604963134 326 / 578 The train loss 0.421763858598 327 / 578 The train loss 0.421067264199 328 / 578 The train loss 0.420072936845 329 / 578 The train loss 0.419251731696 330 / 578 The train loss 0.418379023857 331 / 578 The train loss 0.417438168861 332 / 578 The train loss 0.41677360493 333 / 578 The train loss 0.416361320892 334 / 578 The train loss 0.415483206605 335 / 578 The train loss 0.415068831556 336 / 578 The train loss 0.414106543226 337 / 578 The train loss 0.413195310393 338 / 578 The train loss 0.412211894664 339 / 578 The train loss 0.41139348698 340 / 578 The train loss 0.410815188667 341 / 578 The train loss 0.409826573722 342 / 578 The train loss 0.408782622363 343 / 578 The train loss 0.407821656865 344 / 578 The train loss 0.407031787153 345 / 578 The train loss 0.406087424817 346 / 578 The train loss 0.405205922842 347 / 578 The train loss 0.404699397539 348 / 578 The train loss 0.404297301059 349 / 578 The train loss 0.40424422472 350 / 578 The train loss 0.403663597197 351 / 578 The train loss 0.403054679821 352 / 578 The train loss 0.402568201159 353 / 578 The train loss 0.402436727384 354 / 578 The train loss 0.40262184562 355 / 578 The train loss 0.403176743259 356 / 578 The train loss 0.403224579032 357 / 578 The train loss 0.402910417666 358 / 578 The train loss 0.403478119756 359 / 578 The train loss 0.40387989963 360 / 578 The train loss 0.404213671019 361 / 578 The train loss 0.404205834235 362 / 578 The train loss 0.404215450327 363 / 578 The train loss 0.404246179736 364 / 578 The train loss 0.403653485562 365 / 578 The train loss 0.402712155843 366 / 578 The train loss 0.402096182907 367 / 578 The train loss 0.401186290799 368 / 578 The train loss 0.400274171094 369 / 578 The train loss 0.399427450189 370 / 578 The train loss 0.398761954969 371 / 578 The train loss 0.398114576479 372 / 578 The train loss 0.397349608987 373 / 578 The train loss 0.39665839498 374 / 578 The train loss 0.396058662834 375 / 578 The train loss 0.395552614118 376 / 578 The train loss 0.395212727723 377 / 578 The train loss 0.394564205425 378 / 578 The train loss 0.393984759714 379 / 578 The train loss 0.393118625055 380 / 578 The train loss 0.392334773083 381 / 578 The train loss 0.391913258893 382 / 578 The train loss 0.391768723185 383 / 578 The train loss 0.391343929326 384 / 578 The train loss 0.390610533073 385 / 578 The train loss 0.389784922465 386 / 578 The train loss 0.389209761796 387 / 578 The train loss 0.388549007961 388 / 578 The train loss 0.387949501966 389 / 578 The train loss 0.38723041177 390 / 578 The train loss 0.386595773874 391 / 578 The train loss 0.385839619243 392 / 578 The train loss 0.385062320277 393 / 578 The train loss 0.384190180521 394 / 578 The train loss 0.383344701595 395 / 578 The train loss 0.382465804646 396 / 578 The train loss 0.381869553885 397 / 578 The train loss 0.381247117857 398 / 578 The train loss 0.380602091724 399 / 578 The train loss 0.379939388797 400 / 578 The train loss 0.379175778111 401 / 578 The train loss 0.378395464973 402 / 578 The train loss 0.377651215438 403 / 578 The train loss 0.376900068034 404 / 578 The train loss 0.376158474626 405 / 578 The train loss 0.375488745913 406 / 578 The train loss 0.374878402419 407 / 578 The train loss 0.37408643923 408 / 578 The train loss 0.373475853687 409 / 578 The train loss 0.374620825822 410 / 578 The train loss 0.376488429131 411 / 578 The train loss 0.377536166147 412 / 578 The train loss 0.377639400262 413 / 578 The train loss 0.377065869535 414 / 578 The train loss 0.376593370661 415 / 578 The train loss 0.376157604252 416 / 578 The train loss 0.375425557395 417 / 578 The train loss 0.374807540341 418 / 578 The train loss 0.374083228378 419 / 578 The train loss 0.373408493895 420 / 578 The train loss 0.372650667258 421 / 578 The train loss 0.37204416609 422 / 578 The train loss 0.371624880453 423 / 578 The train loss 0.371307104223 424 / 578 The train loss 0.370600391562 425 / 578 The train loss 0.369938192994 426 / 578 The train loss 0.369282567441 427 / 578 The train loss 0.36865981359 428 / 578 The train loss 0.367949693433 429 / 578 The train loss 0.367283533916 430 / 578 The train loss 0.366474062252 431 / 578 The train loss 0.365693008951 432 / 578 The train loss 0.36499759927 433 / 578 The train loss 0.364606268786 434 / 578 The train loss 0.364906115977 435 / 578 The train loss 0.364987794451 436 / 578 The train loss 0.364913420696 437 / 578 The train loss 0.364403903241 438 / 578 The train loss 0.363738247704 439 / 578 The train loss 0.36323287212 440 / 578 The train loss 0.362700574655 441 / 578 The train loss 0.36204829423 442 / 578 The train loss 0.361348723154 443 / 578 The train loss 0.36059705771 444 / 578 The train loss 0.360200229655 445 / 578 The train loss 0.360166361283 446 / 578 The train loss 0.36017664765 447 / 578 The train loss 0.35994709684 448 / 578 The train loss 0.360002816314 449 / 578 The train loss 0.360524565371 450 / 578 The train loss 0.360615076679 451 / 578 The train loss 0.360393158009 452 / 578 The train loss 0.360113916268 453 / 578 The train loss 0.359822230516 454 / 578 The train loss 0.359312546403 455 / 578 The train loss 0.358650462387 456 / 578 The train loss 0.358081114911 457 / 578 The train loss 0.357502364168 458 / 578 The train loss 0.356909523763 459 / 578 The train loss 0.356222764778 460 / 578 The train loss 0.355562390476 461 / 578 The train loss 0.354894468427 462 / 578 The train loss 0.3542976884 463 / 578 The train loss 0.354946699581 464 / 578 The train loss 0.35589044774 465 / 578 The train loss 0.355657366287 466 / 578 The train loss 0.355133635436 467 / 578 The train loss 0.354511804879 468 / 578 The train loss 0.353819792609 469 / 578 The train loss 0.353257905855 470 / 578 The train loss 0.352810321554 471 / 578 The train loss 0.352326511589 472 / 578 The train loss 0.351890173958 473 / 578 The train loss 0.351335787933 474 / 578 The train loss 0.350797032973 475 / 578 The train loss 0.350291037979 476 / 578 The train loss 0.349732801893 477 / 578 The train loss 0.349238452152 478 / 578 The train loss 0.348809986165 479 / 578 The train loss 0.348431288181 480 / 578 The train loss 0.348055384879 481 / 578 The train loss 0.347858479711 482 / 578 The train loss 0.347616998344 483 / 578 The train loss 0.347235693614 484 / 578 The train loss 0.346909631302 485 / 578 The train loss 0.346440097548 486 / 578 The train loss 0.345915350921 487 / 578 The train loss 0.345541561191 488 / 578 The train loss 0.34530034954 489 / 578 The train loss 0.344956506733 490 / 578 The train loss 0.344733529306 491 / 578 The train loss 0.344368431691 492 / 578 The train loss 0.343969332156 493 / 578 The train loss 0.343795757956 494 / 578 The train loss 0.343575207747 495 / 578 The train loss 0.343290157176 496 / 578 The train loss 0.343392455316 497 / 578 The train loss 0.34393034953 498 / 578 The train loss 0.34408550763 499 / 578 The train loss 0.344087993599 500 / 578 The train loss 0.343899484958 501 / 578 The train loss 0.343564050428 502 / 578 The train loss 0.343245489749 503 / 578 The train loss 0.342915153079 504 / 578 The train loss 0.342638944139 505 / 578 The train loss 0.34234165304 506 / 578 The train loss 0.341934056942 507 / 578 The train loss 0.341571856441 508 / 578 The train loss 0.341083925976 509 / 578 The train loss 0.340568807799 510 / 578 The train loss 0.340141213652 511 / 578 The train loss 0.339719866073 512 / 578 The train loss 0.339418349915 513 / 578 The train loss 0.339106212419 514 / 578 The train loss 0.338636809746 515 / 578 The train loss 0.338384698342 516 / 578 The train loss 0.338285341942 517 / 578 The train loss 0.338032199949 518 / 578 The train loss 0.337647436694 519 / 578 The train loss 0.337283780719 520 / 578 The train loss 0.336995504506 521 / 578 The train loss 0.336697003659 522 / 578 The train loss 0.336215800991 523 / 578 The train loss 0.335782453215 524 / 578 The train loss 0.335441378518 525 / 578 The train loss 0.335156901163 526 / 578 The train loss 0.335049963719 527 / 578 The train loss 0.334877212198 528 / 578 The train loss 0.334568651023 529 / 578 The train loss 0.334405574937 530 / 578 The train loss 0.334177428199 531 / 578 The train loss 0.33399902382 532 / 578 The train loss 0.333913116971 533 / 578 The train loss 0.333938021258 534 / 578 The train loss 0.333726318166 535 / 578 The train loss 0.333235403196 536 / 578 The train loss 0.332758424633 537 / 578 The train loss 0.332430127998 538 / 578 The train loss 0.332172633239 539 / 578 The train loss 0.331811557864 540 / 578 The train loss 0.331381644457 541 / 578 The train loss 0.331050965608 542 / 578 The train loss 0.330778180024 543 / 578 The train loss 0.330453263609 544 / 578 The train loss 0.330263150044 545 / 578 The train loss 0.330440105932 546 / 578 The train loss 0.331836086336 547 / 578 The train loss 0.333623287076 548 / 578 The train loss 0.336751460441 549 / 578 The train loss 0.338608802189 550 / 578 The train loss 0.339729831283 551 / 578 The train loss 0.340646036263 552 / 578 The train loss 0.341362436353 553 / 578 The train loss 0.341936668327 554 / 578 The train loss 0.342492730673 555 / 578 The train loss 0.342993780256 556 / 578 The train loss 0.343460973175 557 / 578 The train loss 0.344446373007 558 / 578 The train loss 0.344927796496 559 / 578 The train loss 0.345118202239 560 / 578 The train loss 0.345006459961 561 / 578 The train loss 0.344691433833 562 / 578 The train loss 0.344326390129 563 / 578 The train loss 0.344152688037 564 / 578 The train loss 0.343875232196 565 / 578 The train loss 0.343517464488 566 / 578 The train loss 0.343123431298 567 / 578 The train loss 0.342703221438 568 / 578 The train loss 0.34221122769 569 / 578 The train loss 0.341651507995 570 / 578 The train loss 0.341134874735 571 / 578 The train loss 0.340742885665 572 / 578 The train loss 0.340327674122 573 / 578 The train loss 0.340446392334 574 / 578 The train loss 0.340654738432 575 / 578 The train loss 0.340656285393 576 / 578 The train loss 0.34064382284 577 / 578 The train loss 0.340558733852 578 / 578 The train loss 0.340436501748 Starting epoch 14 Validation: 1 / 30 The valid loss 0.396561205387 2 / 30 The valid loss 0.367727518082 3 / 30 The valid loss 0.335392594337 4 / 30 The valid loss 0.278526620939 5 / 30 The valid loss 0.247280539572 6 / 30 The valid loss 0.273715010534 7 / 30 The valid loss 0.372606610613 8 / 30 The valid loss 0.50729224924 9 / 30 The valid loss 0.643750145204 10 / 30 The valid loss 0.717086035758 11 / 30 The valid loss 0.742998053404 12 / 30 The valid loss 0.756108970071 13 / 30 The valid loss 0.764925581331 14 / 30 The valid loss 0.77045944812 15 / 30 The valid loss 0.762035088241 16 / 30 The valid loss 0.761598445941 17 / 30 The valid loss 0.776557670358 18 / 30 The valid loss 0.785313245737 19 / 30 The valid loss 0.784621963768 20 / 30 The valid loss 0.781027421728 21 / 30 The valid loss 0.764486199688 22 / 30 The valid loss 0.748506734656 23 / 30 The valid loss 0.735235284528 24 / 30 The valid loss 0.724379973176 25 / 30 The valid loss 0.710756036341 26 / 30 The valid loss 0.700758831146 27 / 30 The valid loss 0.687267992508 28 / 30 The valid loss 0.670311279329 29 / 30 The valid loss 0.654268204909 30 / 30 The valid loss 0.641604083528 Training 1 / 578 The train loss 0.321450829506 2 / 578 The train loss 0.201557390392 3 / 578 The train loss 0.169699475169 4 / 578 The train loss 0.148990418762 5 / 578 The train loss 0.135986748338 6 / 578 The train loss 0.125710392992 7 / 578 The train loss 0.125556617975 8 / 578 The train loss 0.122676471248 9 / 578 The train loss 0.120587726434 10 / 578 The train loss 0.112394492328 11 / 578 The train loss 0.106707094068 12 / 578 The train loss 0.0996844454979 13 / 578 The train loss 0.0960154023308 14 / 578 The train loss 0.105025252593 15 / 578 The train loss 0.109478391707 16 / 578 The train loss 0.117579821963 17 / 578 The train loss 0.127632888363 18 / 578 The train loss 0.135317112423 19 / 578 The train loss 0.139265274139 20 / 578 The train loss 0.136935063824 21 / 578 The train loss 0.134156889149 22 / 578 The train loss 0.130863693119 23 / 578 The train loss 0.128722606956 24 / 578 The train loss 0.125185696874 25 / 578 The train loss 0.121952980161 26 / 578 The train loss 0.118970291259 27 / 578 The train loss 0.116795749024 28 / 578 The train loss 0.113860933376 29 / 578 The train loss 0.11080439742 30 / 578 The train loss 0.108345558432 31 / 578 The train loss 0.105963795836 32 / 578 The train loss 0.103756049008 33 / 578 The train loss 0.102221732395 34 / 578 The train loss 0.100676423103 35 / 578 The train loss 0.100149370996 36 / 578 The train loss 0.100291915911 37 / 578 The train loss 0.100015845502 38 / 578 The train loss 0.0999702582332 39 / 578 The train loss 0.0983730885558 40 / 578 The train loss 0.0968126862776 41 / 578 The train loss 0.0949445594466 42 / 578 The train loss 0.0933965748353 43 / 578 The train loss 0.0928667775507 44 / 578 The train loss 0.0915955612859 45 / 578 The train loss 0.0912421560536 46 / 578 The train loss 0.0904872946684 47 / 578 The train loss 0.0892493621387 48 / 578 The train loss 0.0879898926166 49 / 578 The train loss 0.0872897900717 50 / 578 The train loss 0.0883199968562 51 / 578 The train loss 0.0894217487863 52 / 578 The train loss 0.0899989481371 53 / 578 The train loss 0.0895905769869 54 / 578 The train loss 0.0888229335003 55 / 578 The train loss 0.0878189732744 56 / 578 The train loss 0.0866910674875 57 / 578 The train loss 0.0859154713781 58 / 578 The train loss 0.0865364519173 59 / 578 The train loss 0.0883170328403 60 / 578 The train loss 0.090944716831 61 / 578 The train loss 0.0955599087184 62 / 578 The train loss 0.100297133769 63 / 578 The train loss 0.103497675014 64 / 578 The train loss 0.105620239628 65 / 578 The train loss 0.106584499662 66 / 578 The train loss 0.10599357409 67 / 578 The train loss 0.105038920779 68 / 578 The train loss 0.104899075868 69 / 578 The train loss 0.104133720415 70 / 578 The train loss 0.103133395155 71 / 578 The train loss 0.102190411322 72 / 578 The train loss 0.101234184371 73 / 578 The train loss 0.101126925178 74 / 578 The train loss 0.103359576415 75 / 578 The train loss 0.107101593415 76 / 578 The train loss 0.109537386188 77 / 578 The train loss 0.111153994675 78 / 578 The train loss 0.113414197014 79 / 578 The train loss 0.117407615049 80 / 578 The train loss 0.120437491685 81 / 578 The train loss 0.124012808005 82 / 578 The train loss 0.126711144316 83 / 578 The train loss 0.128664331982 84 / 578 The train loss 0.131537412249 85 / 578 The train loss 0.134432643301 86 / 578 The train loss 0.13755463237 87 / 578 The train loss 0.139818208656 88 / 578 The train loss 0.142205566845 89 / 578 The train loss 0.148848094967 90 / 578 The train loss 0.152773697509 91 / 578 The train loss 0.152388324613 92 / 578 The train loss 0.153012111621 93 / 578 The train loss 0.152984973724 94 / 578 The train loss 0.152893078454 95 / 578 The train loss 0.152901367922 96 / 578 The train loss 0.153349890218 97 / 578 The train loss 0.154469371303 98 / 578 The train loss 0.155817795621 99 / 578 The train loss 0.15715395757 100 / 578 The train loss 0.158770276457 101 / 578 The train loss 0.161079398329 102 / 578 The train loss 0.161052117745 103 / 578 The train loss 0.160452266775 104 / 578 The train loss 0.159819207512 105 / 578 The train loss 0.159391455778 106 / 578 The train loss 0.15976005645 107 / 578 The train loss 0.159977060821 108 / 578 The train loss 0.160298296078 109 / 578 The train loss 0.16052597768 110 / 578 The train loss 0.16041344113 111 / 578 The train loss 0.159995755768 112 / 578 The train loss 0.159603269704 113 / 578 The train loss 0.160463644538 114 / 578 The train loss 0.162108741766 115 / 578 The train loss 0.164123261947 116 / 578 The train loss 0.167831399253 117 / 578 The train loss 0.170248337727 118 / 578 The train loss 0.173323384723 119 / 578 The train loss 0.176825564887 120 / 578 The train loss 0.180263696052 121 / 578 The train loss 0.183750950539 122 / 578 The train loss 0.187886222892 123 / 578 The train loss 0.193447347518 124 / 578 The train loss 0.202338411863 125 / 578 The train loss 0.210681478441 126 / 578 The train loss 0.217185669416 127 / 578 The train loss 0.225378297387 128 / 578 The train loss 0.233871039876 129 / 578 The train loss 0.241870976176 130 / 578 The train loss 0.250456620122 131 / 578 The train loss 0.257956456535 132 / 578 The train loss 0.261571384233 133 / 578 The train loss 0.263013283281 134 / 578 The train loss 0.263442369039 135 / 578 The train loss 0.26454145373 136 / 578 The train loss 0.266744996202 137 / 578 The train loss 0.269035032576 138 / 578 The train loss 0.270183700528 139 / 578 The train loss 0.270503430004 140 / 578 The train loss 0.269658200177 141 / 578 The train loss 0.268629397906 142 / 578 The train loss 0.26857449639 143 / 578 The train loss 0.269301049776 144 / 578 The train loss 0.269338981559 145 / 578 The train loss 0.268866628614 146 / 578 The train loss 0.268873773209 147 / 578 The train loss 0.268763187386 148 / 578 The train loss 0.268295682262 149 / 578 The train loss 0.267829812913 150 / 578 The train loss 0.267142246068 151 / 578 The train loss 0.266134146823 152 / 578 The train loss 0.265354897925 153 / 578 The train loss 0.264794105408 154 / 578 The train loss 0.263893925041 155 / 578 The train loss 0.263408959393 156 / 578 The train loss 0.262679611548 157 / 578 The train loss 0.262443454118 158 / 578 The train loss 0.261767883561 159 / 578 The train loss 0.260692430274 160 / 578 The train loss 0.259987084009 161 / 578 The train loss 0.259499376605 162 / 578 The train loss 0.258695925367 163 / 578 The train loss 0.258009480644 164 / 578 The train loss 0.25683301869 165 / 578 The train loss 0.255769511741 166 / 578 The train loss 0.254838287785 167 / 578 The train loss 0.253991328804 168 / 578 The train loss 0.25721766813 169 / 578 The train loss 0.26049464531 170 / 578 The train loss 0.263795521461 171 / 578 The train loss 0.266306851125 172 / 578 The train loss 0.268685401656 173 / 578 The train loss 0.272010187322 174 / 578 The train loss 0.275192715784 175 / 578 The train loss 0.278382012972 176 / 578 The train loss 0.28080516118 177 / 578 The train loss 0.282745334178 178 / 578 The train loss 0.284818895203 179 / 578 The train loss 0.286486622799 180 / 578 The train loss 0.28758379134 181 / 578 The train loss 0.288226101568 182 / 578 The train loss 0.28811479102 183 / 578 The train loss 0.287517724101 184 / 578 The train loss 0.288750320632 185 / 578 The train loss 0.292362422516 186 / 578 The train loss 0.293848650069 187 / 578 The train loss 0.29575384711 188 / 578 The train loss 0.301426442023 189 / 578 The train loss 0.310258181914 190 / 578 The train loss 0.319636823276 191 / 578 The train loss 0.32537049176 192 / 578 The train loss 0.32996557618 193 / 578 The train loss 0.334188678374 194 / 578 The train loss 0.336650702718 195 / 578 The train loss 0.336476248006 196 / 578 The train loss 0.336798253526 197 / 578 The train loss 0.338299777113 198 / 578 The train loss 0.340814063457 199 / 578 The train loss 0.344506643937 200 / 578 The train loss 0.348536721878 201 / 578 The train loss 0.351303435195 202 / 578 The train loss 0.355159648685 203 / 578 The train loss 0.358573462686 204 / 578 The train loss 0.360790954611 205 / 578 The train loss 0.360366994106 206 / 578 The train loss 0.360782180994 207 / 578 The train loss 0.361425890115 208 / 578 The train loss 0.361615603037 209 / 578 The train loss 0.361725791041 210 / 578 The train loss 0.361592251453 211 / 578 The train loss 0.361293169524 212 / 578 The train loss 0.360382355574 213 / 578 The train loss 0.359665291515 214 / 578 The train loss 0.359235013861 215 / 578 The train loss 0.359202260299 216 / 578 The train loss 0.358335517243 217 / 578 The train loss 0.357173673809 218 / 578 The train loss 0.356095269015 219 / 578 The train loss 0.355164876097 220 / 578 The train loss 0.35400664444 221 / 578 The train loss 0.353044755217 222 / 578 The train loss 0.35259040388 223 / 578 The train loss 0.352698526876 224 / 578 The train loss 0.353049233589 225 / 578 The train loss 0.353297888603 226 / 578 The train loss 0.353703048014 227 / 578 The train loss 0.354285850169 228 / 578 The train loss 0.354758356387 229 / 578 The train loss 0.354794097682 230 / 578 The train loss 0.354848108376 231 / 578 The train loss 0.354448190015 232 / 578 The train loss 0.353945132634 233 / 578 The train loss 0.353982757249 234 / 578 The train loss 0.353926476027 235 / 578 The train loss 0.353659375075 236 / 578 The train loss 0.352729719037 237 / 578 The train loss 0.352855789731 238 / 578 The train loss 0.356013416962 239 / 578 The train loss 0.358249817651 240 / 578 The train loss 0.358008517915 241 / 578 The train loss 0.356917076462 242 / 578 The train loss 0.356325189252 243 / 578 The train loss 0.356452398401 244 / 578 The train loss 0.356801124076 245 / 578 The train loss 0.357152630906 246 / 578 The train loss 0.357500108822 247 / 578 The train loss 0.356491196614 248 / 578 The train loss 0.355659003219 249 / 578 The train loss 0.354953375626 250 / 578 The train loss 0.354957784832 251 / 578 The train loss 0.355830063561 252 / 578 The train loss 0.357138932638 253 / 578 The train loss 0.359296840582 254 / 578 The train loss 0.361932123212 255 / 578 The train loss 0.364099805904 256 / 578 The train loss 0.366976825928 257 / 578 The train loss 0.370705811836 258 / 578 The train loss 0.373356046545 259 / 578 The train loss 0.374358312906 260 / 578 The train loss 0.374970698529 261 / 578 The train loss 0.374783312966 262 / 578 The train loss 0.374671872051 263 / 578 The train loss 0.374491352942 264 / 578 The train loss 0.374329203966 265 / 578 The train loss 0.374397912284 266 / 578 The train loss 0.374503568543 267 / 578 The train loss 0.374469315012 268 / 578 The train loss 0.374082648876 269 / 578 The train loss 0.374040026481 270 / 578 The train loss 0.375396246215 271 / 578 The train loss 0.37561934752 272 / 578 The train loss 0.37502494416 273 / 578 The train loss 0.374688155867 274 / 578 The train loss 0.374030596855 275 / 578 The train loss 0.372954615002 276 / 578 The train loss 0.372014904179 277 / 578 The train loss 0.371167832468 278 / 578 The train loss 0.370682550661 279 / 578 The train loss 0.369721837769 280 / 578 The train loss 0.36975961293 281 / 578 The train loss 0.369310869202 282 / 578 The train loss 0.368543025635 283 / 578 The train loss 0.368549458735 284 / 578 The train loss 0.367881184129 285 / 578 The train loss 0.367405992982 286 / 578 The train loss 0.366629488688 287 / 578 The train loss 0.366045691488 288 / 578 The train loss 0.365636319258 289 / 578 The train loss 0.365442955808 290 / 578 The train loss 0.365238448387 291 / 578 The train loss 0.365218743451 292 / 578 The train loss 0.36511081327 293 / 578 The train loss 0.364666517862 294 / 578 The train loss 0.364063519072 295 / 578 The train loss 0.363622504094 296 / 578 The train loss 0.362889908028 297 / 578 The train loss 0.362135227356 298 / 578 The train loss 0.361354448976 299 / 578 The train loss 0.360921416668 300 / 578 The train loss 0.360951127484 301 / 578 The train loss 0.360996573842 302 / 578 The train loss 0.361173682828 303 / 578 The train loss 0.361375628599 304 / 578 The train loss 0.361218807018 305 / 578 The train loss 0.361403356004 306 / 578 The train loss 0.361678744916 307 / 578 The train loss 0.361523270486 308 / 578 The train loss 0.361757865659 309 / 578 The train loss 0.361702490176 310 / 578 The train loss 0.361314891303 311 / 578 The train loss 0.360769405458 312 / 578 The train loss 0.360567914967 313 / 578 The train loss 0.360187700262 314 / 578 The train loss 0.359508198324 315 / 578 The train loss 0.358640552963 316 / 578 The train loss 0.358869766392 317 / 578 The train loss 0.359459390094 318 / 578 The train loss 0.359592397516 319 / 578 The train loss 0.359653747264 320 / 578 The train loss 0.359537992231 321 / 578 The train loss 0.359143141297 322 / 578 The train loss 0.358747952404 323 / 578 The train loss 0.358504413659 324 / 578 The train loss 0.357640649847 325 / 578 The train loss 0.3570745674 326 / 578 The train loss 0.356623874714 327 / 578 The train loss 0.355718403143 328 / 578 The train loss 0.354704779266 329 / 578 The train loss 0.353752760098 330 / 578 The train loss 0.352738908469 331 / 578 The train loss 0.351828014298 332 / 578 The train loss 0.350926517167 333 / 578 The train loss 0.350004151167 334 / 578 The train loss 0.349208972495 335 / 578 The train loss 0.348402538712 336 / 578 The train loss 0.348072657533 337 / 578 The train loss 0.347647305274 338 / 578 The train loss 0.347171950705 339 / 578 The train loss 0.346740883091 340 / 578 The train loss 0.345961193297 341 / 578 The train loss 0.345041952902 342 / 578 The train loss 0.344152368022 343 / 578 The train loss 0.343399852085 344 / 578 The train loss 0.342613036468 345 / 578 The train loss 0.3419352533 346 / 578 The train loss 0.341234432442 347 / 578 The train loss 0.340877476409 348 / 578 The train loss 0.340631272478 349 / 578 The train loss 0.340402310981 350 / 578 The train loss 0.339844416981 351 / 578 The train loss 0.339337442681 352 / 578 The train loss 0.338897518988 353 / 578 The train loss 0.338381739587 354 / 578 The train loss 0.338159327206 355 / 578 The train loss 0.338577085823 356 / 578 The train loss 0.338540390558 357 / 578 The train loss 0.338572186337 358 / 578 The train loss 0.338384266261 359 / 578 The train loss 0.338070096043 360 / 578 The train loss 0.338299802572 361 / 578 The train loss 0.338572868314 362 / 578 The train loss 0.338382127664 363 / 578 The train loss 0.338145778163 364 / 578 The train loss 0.337865227777 365 / 578 The train loss 0.337388765245 366 / 578 The train loss 0.3366120041 367 / 578 The train loss 0.335870098236 368 / 578 The train loss 0.335102398693 369 / 578 The train loss 0.334286404309 370 / 578 The train loss 0.333512793436 371 / 578 The train loss 0.332710461932 372 / 578 The train loss 0.331938770615 373 / 578 The train loss 0.331226286581 374 / 578 The train loss 0.330560454616 375 / 578 The train loss 0.33005034326 376 / 578 The train loss 0.329391854067 377 / 578 The train loss 0.328591802274 378 / 578 The train loss 0.327805618278 379 / 578 The train loss 0.327085008221 380 / 578 The train loss 0.326388422802 381 / 578 The train loss 0.325954606771 382 / 578 The train loss 0.32587378867 383 / 578 The train loss 0.325589660613 384 / 578 The train loss 0.325163139777 385 / 578 The train loss 0.324793111823 386 / 578 The train loss 0.324253982879 387 / 578 The train loss 0.323625924581 388 / 578 The train loss 0.322930801929 389 / 578 The train loss 0.322327945295 390 / 578 The train loss 0.321565677278 391 / 578 The train loss 0.320806034753 392 / 578 The train loss 0.320095572796 393 / 578 The train loss 0.319516879822 394 / 578 The train loss 0.319075776902 395 / 578 The train loss 0.319030943612 396 / 578 The train loss 0.318649320313 397 / 578 The train loss 0.318091762086 398 / 578 The train loss 0.317389729957 399 / 578 The train loss 0.316735342084 400 / 578 The train loss 0.31613040132 401 / 578 The train loss 0.315607521725 402 / 578 The train loss 0.315002784332 403 / 578 The train loss 0.314278697669 404 / 578 The train loss 0.313651288597 405 / 578 The train loss 0.313014965113 406 / 578 The train loss 0.312374170379 407 / 578 The train loss 0.311770660235 408 / 578 The train loss 0.311240942532 409 / 578 The train loss 0.312779022098 410 / 578 The train loss 0.314765124573 411 / 578 The train loss 0.31561998795 412 / 578 The train loss 0.315630755093 413 / 578 The train loss 0.315360102329 414 / 578 The train loss 0.315504884235 415 / 578 The train loss 0.31615546181 416 / 578 The train loss 0.316121945234 417 / 578 The train loss 0.315734607786 418 / 578 The train loss 0.315140738446 419 / 578 The train loss 0.314497467508 420 / 578 The train loss 0.313855443899 421 / 578 The train loss 0.313283278251 422 / 578 The train loss 0.312913432633 423 / 578 The train loss 0.312570616167 424 / 578 The train loss 0.312225410315 425 / 578 The train loss 0.312052921464 426 / 578 The train loss 0.311940364352 427 / 578 The train loss 0.312215024317 428 / 578 The train loss 0.31225612055 429 / 578 The train loss 0.311940909913 430 / 578 The train loss 0.311579881838 431 / 578 The train loss 0.310994528722 432 / 578 The train loss 0.310323237209 433 / 578 The train loss 0.30979777722 434 / 578 The train loss 0.309836439385 435 / 578 The train loss 0.309797788874 436 / 578 The train loss 0.309437415443 437 / 578 The train loss 0.309019835824 438 / 578 The train loss 0.308621859824 439 / 578 The train loss 0.308306318863 440 / 578 The train loss 0.30806088699 441 / 578 The train loss 0.307743820955 442 / 578 The train loss 0.307216495149 443 / 578 The train loss 0.306765992783 444 / 578 The train loss 0.306616217645 445 / 578 The train loss 0.307006324618 446 / 578 The train loss 0.307600223773 447 / 578 The train loss 0.307753087232 448 / 578 The train loss 0.307808101718 449 / 578 The train loss 0.307749191287 450 / 578 The train loss 0.307584733607 451 / 578 The train loss 0.307509087976 452 / 578 The train loss 0.307448973278 453 / 578 The train loss 0.306991681021 454 / 578 The train loss 0.306511145635 455 / 578 The train loss 0.306139542351 456 / 578 The train loss 0.305721191077 457 / 578 The train loss 0.305455562167 458 / 578 The train loss 0.305058424349 459 / 578 The train loss 0.304555193556 460 / 578 The train loss 0.303989133368 461 / 578 The train loss 0.303532137513 462 / 578 The train loss 0.303088226811 463 / 578 The train loss 0.303602244439 464 / 578 The train loss 0.304044931549 465 / 578 The train loss 0.304050740632 466 / 578 The train loss 0.303830525394 467 / 578 The train loss 0.303366637125 468 / 578 The train loss 0.303136127738 469 / 578 The train loss 0.303234873661 470 / 578 The train loss 0.303145226757 471 / 578 The train loss 0.302695734296 472 / 578 The train loss 0.302294474758 473 / 578 The train loss 0.301939215739 474 / 578 The train loss 0.301527730712 475 / 578 The train loss 0.301064049006 476 / 578 The train loss 0.300576548319 477 / 578 The train loss 0.300083112214 478 / 578 The train loss 0.299653593731 479 / 578 The train loss 0.299236028344 480 / 578 The train loss 0.298950989249 481 / 578 The train loss 0.29874662536 482 / 578 The train loss 0.298675243275 483 / 578 The train loss 0.298436412576 484 / 578 The train loss 0.29820029319 485 / 578 The train loss 0.297935742524 486 / 578 The train loss 0.297532623917 487 / 578 The train loss 0.297320950922 488 / 578 The train loss 0.297069237842 489 / 578 The train loss 0.296840905671 490 / 578 The train loss 0.296792325332 491 / 578 The train loss 0.296867552834 492 / 578 The train loss 0.296770413005 493 / 578 The train loss 0.296557743261 494 / 578 The train loss 0.296235885441 495 / 578 The train loss 0.295992233732 496 / 578 The train loss 0.295942548918 497 / 578 The train loss 0.296256169782 498 / 578 The train loss 0.296650391581 499 / 578 The train loss 0.296742241061 500 / 578 The train loss 0.296488373086 501 / 578 The train loss 0.29619671156 502 / 578 The train loss 0.295822314606 503 / 578 The train loss 0.295500062939 504 / 578 The train loss 0.295352099242 505 / 578 The train loss 0.295027625487 506 / 578 The train loss 0.294666786969 507 / 578 The train loss 0.29431409183 508 / 578 The train loss 0.293847993896 509 / 578 The train loss 0.293434764202 510 / 578 The train loss 0.29303077135 511 / 578 The train loss 0.292601865478 512 / 578 The train loss 0.292246922603 513 / 578 The train loss 0.291983066813 514 / 578 The train loss 0.291713893718 515 / 578 The train loss 0.291494234485 516 / 578 The train loss 0.291197635854 517 / 578 The train loss 0.291067946306 518 / 578 The train loss 0.291003293527 519 / 578 The train loss 0.290859175225 520 / 578 The train loss 0.291100417264 521 / 578 The train loss 0.291033635913 522 / 578 The train loss 0.29061433543 523 / 578 The train loss 0.290318018805 524 / 578 The train loss 0.290063752089 525 / 578 The train loss 0.28988588695 526 / 578 The train loss 0.289541170454 527 / 578 The train loss 0.289330475442 528 / 578 The train loss 0.289224422198 529 / 578 The train loss 0.289033114713 530 / 578 The train loss 0.28890448266 531 / 578 The train loss 0.288784491926 532 / 578 The train loss 0.288748435228 533 / 578 The train loss 0.28873187639 534 / 578 The train loss 0.288424202607 535 / 578 The train loss 0.288099323366 536 / 578 The train loss 0.287772284876 537 / 578 The train loss 0.287507341866 538 / 578 The train loss 0.287487691912 539 / 578 The train loss 0.287298722286 540 / 578 The train loss 0.286980508903 541 / 578 The train loss 0.286766278711 542 / 578 The train loss 0.286676816285 543 / 578 The train loss 0.286349789406 544 / 578 The train loss 0.286232348778 545 / 578 The train loss 0.286384040418 546 / 578 The train loss 0.287417031934 547 / 578 The train loss 0.289337810939 548 / 578 The train loss 0.29267980346 549 / 578 The train loss 0.294493295266 550 / 578 The train loss 0.295640710294 551 / 578 The train loss 0.296343014401 552 / 578 The train loss 0.29679326813 553 / 578 The train loss 0.297378750676 554 / 578 The train loss 0.297730536856 555 / 578 The train loss 0.297833429022 556 / 578 The train loss 0.29786489794 557 / 578 The train loss 0.297815437404 558 / 578 The train loss 0.297547429186 559 / 578 The train loss 0.297607793764 560 / 578 The train loss 0.297612208367 561 / 578 The train loss 0.297340530364 562 / 578 The train loss 0.297007898525 563 / 578 The train loss 0.296942166865 564 / 578 The train loss 0.29684992692 565 / 578 The train loss 0.296501288794 566 / 578 The train loss 0.296135923782 567 / 578 The train loss 0.295829222395 568 / 578 The train loss 0.295521427156 569 / 578 The train loss 0.295145846196 570 / 578 The train loss 0.294764462937 571 / 578 The train loss 0.294394136062 572 / 578 The train loss 0.293938924252 573 / 578 The train loss 0.293951277775 574 / 578 The train loss 0.293849405615 575 / 578 The train loss 0.29358403763 576 / 578 The train loss 0.293699123819 577 / 578 The train loss 0.293975525624 578 / 578 The train loss 0.294186117353 Starting epoch 15 Validation: 1 / 30 The valid loss 0.403793215752 2 / 30 The valid loss 0.390434771776 3 / 30 The valid loss 0.359965632359 4 / 30 The valid loss 0.288358191028 5 / 30 The valid loss 0.24877409339 6 / 30 The valid loss 0.274084508419 7 / 30 The valid loss 0.37223477023 8 / 30 The valid loss 0.510747417808 9 / 30 The valid loss 0.650572339694 10 / 30 The valid loss 0.723547029495 11 / 30 The valid loss 0.765079725872 12 / 30 The valid loss 0.786363164584 13 / 30 The valid loss 0.799053182969 14 / 30 The valid loss 0.808880303587 15 / 30 The valid loss 0.803000879288 16 / 30 The valid loss 0.805204946548 17 / 30 The valid loss 0.824568808079 18 / 30 The valid loss 0.83740904265 19 / 30 The valid loss 0.83839165537 20 / 30 The valid loss 0.834356185794 21 / 30 The valid loss 0.81683722848 22 / 30 The valid loss 0.798889272592 23 / 30 The valid loss 0.783811323021 24 / 30 The valid loss 0.771138694137 25 / 30 The valid loss 0.755037533045 26 / 30 The valid loss 0.742494295423 27 / 30 The valid loss 0.72485341518 28 / 30 The valid loss 0.705200816904 29 / 30 The valid loss 0.688617017249 30 / 30 The valid loss 0.676348614196 Training 1 / 578 The train loss 0.379775941372 2 / 578 The train loss 0.245047815144 3 / 578 The train loss 0.218325520555 4 / 578 The train loss 0.240250576288 5 / 578 The train loss 0.249354621768 6 / 578 The train loss 0.239347736041 7 / 578 The train loss 0.233151242137 8 / 578 The train loss 0.228250725195 9 / 578 The train loss 0.216273170378 10 / 578 The train loss 0.2095610708 11 / 578 The train loss 0.20709211447 12 / 578 The train loss 0.201133703192 13 / 578 The train loss 0.188111554545 14 / 578 The train loss 0.186650243721 15 / 578 The train loss 0.186145270367 16 / 578 The train loss 0.184806599747 17 / 578 The train loss 0.185441788943 18 / 578 The train loss 0.187362302095 19 / 578 The train loss 0.186908332533 20 / 578 The train loss 0.180926799029 21 / 578 The train loss 0.173468187274 22 / 578 The train loss 0.167806106874 23 / 578 The train loss 0.162326182684 24 / 578 The train loss 0.15704724829 25 / 578 The train loss 0.152428629249 26 / 578 The train loss 0.14792177425 27 / 578 The train loss 0.144221957911 28 / 578 The train loss 0.140079865897 29 / 578 The train loss 0.136095682994 30 / 578 The train loss 0.132447094284 31 / 578 The train loss 0.129419810169 32 / 578 The train loss 0.126179150073 33 / 578 The train loss 0.123492919473 34 / 578 The train loss 0.120389758883 35 / 578 The train loss 0.119136603602 36 / 578 The train loss 0.119019394327 37 / 578 The train loss 0.118930169658 38 / 578 The train loss 0.117961888054 39 / 578 The train loss 0.115569104608 40 / 578 The train loss 0.113533377787 41 / 578 The train loss 0.110969019191 42 / 578 The train loss 0.108789683275 43 / 578 The train loss 0.107356918586 44 / 578 The train loss 0.106091903777 45 / 578 The train loss 0.10534304339 46 / 578 The train loss 0.104297451108 47 / 578 The train loss 0.102750538511 48 / 578 The train loss 0.102049765643 49 / 578 The train loss 0.101589715602 50 / 578 The train loss 0.102805202603 51 / 578 The train loss 0.102756932816 52 / 578 The train loss 0.103503624264 53 / 578 The train loss 0.103051960328 54 / 578 The train loss 0.102120614576 55 / 578 The train loss 0.101055692001 56 / 578 The train loss 0.0997328596589 57 / 578 The train loss 0.0989493810033 58 / 578 The train loss 0.100152307158 59 / 578 The train loss 0.101645714272 60 / 578 The train loss 0.104386460409 61 / 578 The train loss 0.10927657722 62 / 578 The train loss 0.116047776635 63 / 578 The train loss 0.120106843374 64 / 578 The train loss 0.122957142186 65 / 578 The train loss 0.124456454584 66 / 578 The train loss 0.123528654264 67 / 578 The train loss 0.122373652825 68 / 578 The train loss 0.121822023819 69 / 578 The train loss 0.120682898909 70 / 578 The train loss 0.119697547544 71 / 578 The train loss 0.119200745571 72 / 578 The train loss 0.119122032542 73 / 578 The train loss 0.118867144958 74 / 578 The train loss 0.121611175052 75 / 578 The train loss 0.128078720917 76 / 578 The train loss 0.130792415799 77 / 578 The train loss 0.131729269521 78 / 578 The train loss 0.132854065213 79 / 578 The train loss 0.135276263792 80 / 578 The train loss 0.136997671193 81 / 578 The train loss 0.13930958928 82 / 578 The train loss 0.140802327225 83 / 578 The train loss 0.142800818141 84 / 578 The train loss 0.145643992926 85 / 578 The train loss 0.146989496447 86 / 578 The train loss 0.147730597706 87 / 578 The train loss 0.150084038704 88 / 578 The train loss 0.152499263386 89 / 578 The train loss 0.157429927646 90 / 578 The train loss 0.160113558297 91 / 578 The train loss 0.159894290002 92 / 578 The train loss 0.160287199582 93 / 578 The train loss 0.159994462285 94 / 578 The train loss 0.160519722572 95 / 578 The train loss 0.161428071284 96 / 578 The train loss 0.161297526172 97 / 578 The train loss 0.161491314737 98 / 578 The train loss 0.161480892792 99 / 578 The train loss 0.161476539203 100 / 578 The train loss 0.162694377862 101 / 578 The train loss 0.162930934257 102 / 578 The train loss 0.162040864245 103 / 578 The train loss 0.161016410524 104 / 578 The train loss 0.159882554605 105 / 578 The train loss 0.158754501492 106 / 578 The train loss 0.158138909254 107 / 578 The train loss 0.157586957215 108 / 578 The train loss 0.156741352648 109 / 578 The train loss 0.156538390478 110 / 578 The train loss 0.156718521599 111 / 578 The train loss 0.157304761747 112 / 578 The train loss 0.156874800866 113 / 578 The train loss 0.156304350304 114 / 578 The train loss 0.155876382891 115 / 578 The train loss 0.155485988797 116 / 578 The train loss 0.155430066528 117 / 578 The train loss 0.155699024693 118 / 578 The train loss 0.156526824523 119 / 578 The train loss 0.157627320434 120 / 578 The train loss 0.159081471805 121 / 578 The train loss 0.160216400785 122 / 578 The train loss 0.162121320449 123 / 578 The train loss 0.165024537472 124 / 578 The train loss 0.171693469338 125 / 578 The train loss 0.178481969923 126 / 578 The train loss 0.183709949344 127 / 578 The train loss 0.188776002244 128 / 578 The train loss 0.194982109679 129 / 578 The train loss 0.202173354176 130 / 578 The train loss 0.210111020248 131 / 578 The train loss 0.215161590261 132 / 578 The train loss 0.21693253187 133 / 578 The train loss 0.217675808089 134 / 578 The train loss 0.217456883987 135 / 578 The train loss 0.217962747674 136 / 578 The train loss 0.219325628871 137 / 578 The train loss 0.221232372125 138 / 578 The train loss 0.221525781909 139 / 578 The train loss 0.221209195422 140 / 578 The train loss 0.220949028619 141 / 578 The train loss 0.220657221048 142 / 578 The train loss 0.219860520293 143 / 578 The train loss 0.220149881278 144 / 578 The train loss 0.220042211547 145 / 578 The train loss 0.219898368393 146 / 578 The train loss 0.220386747449 147 / 578 The train loss 0.22075557914 148 / 578 The train loss 0.220841841143 149 / 578 The train loss 0.220989830567 150 / 578 The train loss 0.22119079036 151 / 578 The train loss 0.22111650396 152 / 578 The train loss 0.220879091857 153 / 578 The train loss 0.2206722393 154 / 578 The train loss 0.220310242653 155 / 578 The train loss 0.22111933128 156 / 578 The train loss 0.221613568111 157 / 578 The train loss 0.221769536306 158 / 578 The train loss 0.222009268483 159 / 578 The train loss 0.221669674787 160 / 578 The train loss 0.220887673623 161 / 578 The train loss 0.220326281274 162 / 578 The train loss 0.220004787025 163 / 578 The train loss 0.221004804745 164 / 578 The train loss 0.220600768538 165 / 578 The train loss 0.220131323712 166 / 578 The train loss 0.219722837008 167 / 578 The train loss 0.219543749857 168 / 578 The train loss 0.223154977119 169 / 578 The train loss 0.227043656004 170 / 578 The train loss 0.230463269682 171 / 578 The train loss 0.232755660815 172 / 578 The train loss 0.235111217318 173 / 578 The train loss 0.23753863347 174 / 578 The train loss 0.239419628299 175 / 578 The train loss 0.240879373018 176 / 578 The train loss 0.242782260355 177 / 578 The train loss 0.243453874574 178 / 578 The train loss 0.243852526096 179 / 578 The train loss 0.245326686651 180 / 578 The train loss 0.246692038721 181 / 578 The train loss 0.246907309747 182 / 578 The train loss 0.246421643809 183 / 578 The train loss 0.24660302493 184 / 578 The train loss 0.247676220095 185 / 578 The train loss 0.249526790489 186 / 578 The train loss 0.254348158656 187 / 578 The train loss 0.258490256268 188 / 578 The train loss 0.263369290574 189 / 578 The train loss 0.267882788445 190 / 578 The train loss 0.27368821139 191 / 578 The train loss 0.277724639732 192 / 578 The train loss 0.27947638601 193 / 578 The train loss 0.279544517919 194 / 578 The train loss 0.278781114761 195 / 578 The train loss 0.278851623394 196 / 578 The train loss 0.280989386274 197 / 578 The train loss 0.283567007707 198 / 578 The train loss 0.285490475802 199 / 578 The train loss 0.289683339865 200 / 578 The train loss 0.294478755277 201 / 578 The train loss 0.299945308479 202 / 578 The train loss 0.304780334216 203 / 578 The train loss 0.307409323183 204 / 578 The train loss 0.307641728526 205 / 578 The train loss 0.306847720739 206 / 578 The train loss 0.307236646546 207 / 578 The train loss 0.307885121446 208 / 578 The train loss 0.307647397437 209 / 578 The train loss 0.308143091669 210 / 578 The train loss 0.307945661584 211 / 578 The train loss 0.30717163909 212 / 578 The train loss 0.30609288592 213 / 578 The train loss 0.305561963154 214 / 578 The train loss 0.305273483272 215 / 578 The train loss 0.305626082403 216 / 578 The train loss 0.305844496228 217 / 578 The train loss 0.305414339897 218 / 578 The train loss 0.305514443204 219 / 578 The train loss 0.305459657471 220 / 578 The train loss 0.305329326997 221 / 578 The train loss 0.305171700607 222 / 578 The train loss 0.304983570752 223 / 578 The train loss 0.305063411376 224 / 578 The train loss 0.304990135782 225 / 578 The train loss 0.304960616943 226 / 578 The train loss 0.305231589247 227 / 578 The train loss 0.306351252967 228 / 578 The train loss 0.307030298061 229 / 578 The train loss 0.30784453014 230 / 578 The train loss 0.308751021892 231 / 578 The train loss 0.310034771873 232 / 578 The train loss 0.311466906204 233 / 578 The train loss 0.312577557011 234 / 578 The train loss 0.314138728695 235 / 578 The train loss 0.315147308958 236 / 578 The train loss 0.31541754872 237 / 578 The train loss 0.316252085589 238 / 578 The train loss 0.318567637245 239 / 578 The train loss 0.320046833977 240 / 578 The train loss 0.319650150323 241 / 578 The train loss 0.319124475663 242 / 578 The train loss 0.319590678131 243 / 578 The train loss 0.320555416139 244 / 578 The train loss 0.320502415162 245 / 578 The train loss 0.320209057097 246 / 578 The train loss 0.320120676489 247 / 578 The train loss 0.319176877996 248 / 578 The train loss 0.318385347258 249 / 578 The train loss 0.317719129748 250 / 578 The train loss 0.317827682808 251 / 578 The train loss 0.318669325164 252 / 578 The train loss 0.319611285488 253 / 578 The train loss 0.320899274583 254 / 578 The train loss 0.32291582328 255 / 578 The train loss 0.325396010995 256 / 578 The train loss 0.328097654579 257 / 578 The train loss 0.331396763035 258 / 578 The train loss 0.333455483653 259 / 578 The train loss 0.334782846027 260 / 578 The train loss 0.336196085882 261 / 578 The train loss 0.336651173203 262 / 578 The train loss 0.337399480978 263 / 578 The train loss 0.33763826943 264 / 578 The train loss 0.338299273838 265 / 578 The train loss 0.338740239574 266 / 578 The train loss 0.339673659477 267 / 578 The train loss 0.340046180726 268 / 578 The train loss 0.340508612081 269 / 578 The train loss 0.342107514757 270 / 578 The train loss 0.344947584318 271 / 578 The train loss 0.346444766309 272 / 578 The train loss 0.346361613737 273 / 578 The train loss 0.346281671767 274 / 578 The train loss 0.345923098946 275 / 578 The train loss 0.345158981694 276 / 578 The train loss 0.344314554369 277 / 578 The train loss 0.343271788116 278 / 578 The train loss 0.342468804486 279 / 578 The train loss 0.341613521526 280 / 578 The train loss 0.341119694204 281 / 578 The train loss 0.341016462185 282 / 578 The train loss 0.340713738259 283 / 578 The train loss 0.340437877436 284 / 578 The train loss 0.340274070691 285 / 578 The train loss 0.339709093879 286 / 578 The train loss 0.339288604609 287 / 578 The train loss 0.339117453982 288 / 578 The train loss 0.338740805552 289 / 578 The train loss 0.338217013203 290 / 578 The train loss 0.338094628654 291 / 578 The train loss 0.337726199827 292 / 578 The train loss 0.337161371229 293 / 578 The train loss 0.33676181492 294 / 578 The train loss 0.336252323447 295 / 578 The train loss 0.335597245698 296 / 578 The train loss 0.335008741256 297 / 578 The train loss 0.334356753488 298 / 578 The train loss 0.333647417467 299 / 578 The train loss 0.333150807008 300 / 578 The train loss 0.332921046391 301 / 578 The train loss 0.332923795075 302 / 578 The train loss 0.333564606501 303 / 578 The train loss 0.334178719165 304 / 578 The train loss 0.334320271461 305 / 578 The train loss 0.334496177 306 / 578 The train loss 0.334008589217 307 / 578 The train loss 0.333332073053 308 / 578 The train loss 0.33302654251 309 / 578 The train loss 0.332509447094 310 / 578 The train loss 0.332117684065 311 / 578 The train loss 0.331951782658 312 / 578 The train loss 0.332179630629 313 / 578 The train loss 0.33216936046 314 / 578 The train loss 0.331597281798 315 / 578 The train loss 0.33085604854 316 / 578 The train loss 0.330460626302 317 / 578 The train loss 0.331099712322 318 / 578 The train loss 0.331653184486 319 / 578 The train loss 0.331800083568 320 / 578 The train loss 0.331629148126 321 / 578 The train loss 0.33137414811 322 / 578 The train loss 0.330991879383 323 / 578 The train loss 0.330539854734 324 / 578 The train loss 0.329797776669 325 / 578 The train loss 0.328970743888 326 / 578 The train loss 0.328373157734 327 / 578 The train loss 0.327587501206 328 / 578 The train loss 0.326806581704 329 / 578 The train loss 0.326128633132 330 / 578 The train loss 0.325356226968 331 / 578 The train loss 0.324657103863 332 / 578 The train loss 0.323963453039 333 / 578 The train loss 0.323164681437 334 / 578 The train loss 0.322565026425 335 / 578 The train loss 0.322160929464 336 / 578 The train loss 0.321555456191 337 / 578 The train loss 0.321144177095 338 / 578 The train loss 0.320704038459 339 / 578 The train loss 0.320169608106 340 / 578 The train loss 0.31954942006 341 / 578 The train loss 0.319046696238 342 / 578 The train loss 0.318321576135 343 / 578 The train loss 0.317535662117 344 / 578 The train loss 0.316837627397 345 / 578 The train loss 0.316179672467 346 / 578 The train loss 0.315385755905 347 / 578 The train loss 0.314691341769 348 / 578 The train loss 0.314279435288 349 / 578 The train loss 0.313998400162 350 / 578 The train loss 0.313406846374 351 / 578 The train loss 0.312938241049 352 / 578 The train loss 0.312286298341 353 / 578 The train loss 0.311788087653 354 / 578 The train loss 0.311891895103 355 / 578 The train loss 0.312135889203 356 / 578 The train loss 0.312275370562 357 / 578 The train loss 0.312371802388 358 / 578 The train loss 0.312313166785 359 / 578 The train loss 0.311972122463 360 / 578 The train loss 0.311754399497 361 / 578 The train loss 0.311818793911 362 / 578 The train loss 0.311809563793 363 / 578 The train loss 0.311672557443 364 / 578 The train loss 0.311397923955 365 / 578 The train loss 0.311001320408 366 / 578 The train loss 0.310741742533 367 / 578 The train loss 0.310116138096 368 / 578 The train loss 0.309371640308 369 / 578 The train loss 0.308623006421 370 / 578 The train loss 0.307997569452 371 / 578 The train loss 0.30746948226 372 / 578 The train loss 0.306860702182 373 / 578 The train loss 0.306205017447 374 / 578 The train loss 0.30562297453 375 / 578 The train loss 0.304935075253 376 / 578 The train loss 0.304353615002 377 / 578 The train loss 0.303664848959 378 / 578 The train loss 0.302940249197 379 / 578 The train loss 0.302226177532 380 / 578 The train loss 0.30159731171 381 / 578 The train loss 0.301094526176 382 / 578 The train loss 0.3014217404 383 / 578 The train loss 0.301814613421 384 / 578 The train loss 0.301481051012 385 / 578 The train loss 0.301132922852 386 / 578 The train loss 0.300551368097 387 / 578 The train loss 0.300063644663 388 / 578 The train loss 0.299450537507 389 / 578 The train loss 0.29884585697 390 / 578 The train loss 0.298182876437 391 / 578 The train loss 0.297576979403 392 / 578 The train loss 0.296987073239 393 / 578 The train loss 0.296369427971 394 / 578 The train loss 0.295769795908 395 / 578 The train loss 0.295326511298 396 / 578 The train loss 0.294737173931 397 / 578 The train loss 0.294142930686 398 / 578 The train loss 0.293495181237 399 / 578 The train loss 0.292883883424 400 / 578 The train loss 0.292320715021 401 / 578 The train loss 0.291764044275 402 / 578 The train loss 0.291288797016 403 / 578 The train loss 0.290735506372 404 / 578 The train loss 0.290274260215 405 / 578 The train loss 0.289649472901 406 / 578 The train loss 0.289071328791 407 / 578 The train loss 0.288492338215 408 / 578 The train loss 0.288106490137 409 / 578 The train loss 0.289293418295 410 / 578 The train loss 0.290982421569 411 / 578 The train loss 0.291948943669 412 / 578 The train loss 0.292264756855 413 / 578 The train loss 0.292770254329 414 / 578 The train loss 0.293367101869 415 / 578 The train loss 0.293835660196 416 / 578 The train loss 0.293737764676 417 / 578 The train loss 0.293309841341 418 / 578 The train loss 0.292751508386 419 / 578 The train loss 0.292260939695 420 / 578 The train loss 0.291686523466 421 / 578 The train loss 0.291110013309 422 / 578 The train loss 0.290487577781 423 / 578 The train loss 0.290045293391 424 / 578 The train loss 0.289622499612 425 / 578 The train loss 0.289410623474 426 / 578 The train loss 0.289407462323 427 / 578 The train loss 0.28938642265 428 / 578 The train loss 0.289203935055 429 / 578 The train loss 0.288994635306 430 / 578 The train loss 0.288779128694 431 / 578 The train loss 0.288283140595 432 / 578 The train loss 0.287795774505 433 / 578 The train loss 0.287430727046 434 / 578 The train loss 0.287256549344 435 / 578 The train loss 0.287264562993 436 / 578 The train loss 0.287125907672 437 / 578 The train loss 0.286671533128 438 / 578 The train loss 0.28630280063 439 / 578 The train loss 0.286044694199 440 / 578 The train loss 0.285829970346 441 / 578 The train loss 0.285578200881 442 / 578 The train loss 0.285150695731 443 / 578 The train loss 0.284764954307 444 / 578 The train loss 0.284553512381 445 / 578 The train loss 0.28463715682 446 / 578 The train loss 0.284844843917 447 / 578 The train loss 0.285056388539 448 / 578 The train loss 0.284895940671 449 / 578 The train loss 0.284507066961 450 / 578 The train loss 0.284183336087 451 / 578 The train loss 0.283848033962 452 / 578 The train loss 0.283550966914 453 / 578 The train loss 0.283104801616 454 / 578 The train loss 0.282610866865 455 / 578 The train loss 0.282133918972 456 / 578 The train loss 0.281720852925 457 / 578 The train loss 0.281248409199 458 / 578 The train loss 0.280698882786 459 / 578 The train loss 0.280168493593 460 / 578 The train loss 0.279682056318 461 / 578 The train loss 0.279184654181 462 / 578 The train loss 0.278929248666 463 / 578 The train loss 0.279787316198 464 / 578 The train loss 0.280303084279 465 / 578 The train loss 0.280296906945 466 / 578 The train loss 0.280152681643 467 / 578 The train loss 0.279728443312 468 / 578 The train loss 0.279287597869 469 / 578 The train loss 0.279062868328 470 / 578 The train loss 0.279015941779 471 / 578 The train loss 0.278906761734 472 / 578 The train loss 0.278630484357 473 / 578 The train loss 0.278233018997 474 / 578 The train loss 0.277837281791 475 / 578 The train loss 0.277405499679 476 / 578 The train loss 0.277071442694 477 / 578 The train loss 0.276740137109 478 / 578 The train loss 0.276309009256 479 / 578 The train loss 0.275919744924 480 / 578 The train loss 0.27560229246 481 / 578 The train loss 0.275388324959 482 / 578 The train loss 0.275240515869 483 / 578 The train loss 0.274886994077 484 / 578 The train loss 0.274606080591 485 / 578 The train loss 0.274375084692 486 / 578 The train loss 0.274029665558 487 / 578 The train loss 0.273619282161 488 / 578 The train loss 0.273274311128 489 / 578 The train loss 0.272994440609 490 / 578 The train loss 0.272915975175 491 / 578 The train loss 0.272898794681 492 / 578 The train loss 0.272851663486 493 / 578 The train loss 0.272647292937 494 / 578 The train loss 0.272314781334 495 / 578 The train loss 0.272069485625 496 / 578 The train loss 0.272047839101 497 / 578 The train loss 0.271969046695 498 / 578 The train loss 0.271943075773 499 / 578 The train loss 0.271886107413 500 / 578 The train loss 0.271697971556 501 / 578 The train loss 0.271470102011 502 / 578 The train loss 0.271284041685 503 / 578 The train loss 0.271033369171 504 / 578 The train loss 0.270775553404 505 / 578 The train loss 0.270519881766 506 / 578 The train loss 0.27032197703 507 / 578 The train loss 0.270145841931 508 / 578 The train loss 0.269786725763 509 / 578 The train loss 0.269356192531 510 / 578 The train loss 0.268936546728 511 / 578 The train loss 0.268490506445 512 / 578 The train loss 0.268146002254 513 / 578 The train loss 0.267801711604 514 / 578 The train loss 0.267447451756 515 / 578 The train loss 0.267133995844 516 / 578 The train loss 0.267153878642 517 / 578 The train loss 0.26702460812 518 / 578 The train loss 0.266797198389 519 / 578 The train loss 0.266603273359 520 / 578 The train loss 0.266442148112 521 / 578 The train loss 0.266243375091 522 / 578 The train loss 0.265880621381 523 / 578 The train loss 0.265697657617 524 / 578 The train loss 0.265543236315 525 / 578 The train loss 0.265336630327 526 / 578 The train loss 0.265170555558 527 / 578 The train loss 0.265095308292 528 / 578 The train loss 0.264909649702 529 / 578 The train loss 0.264761699941 530 / 578 The train loss 0.264767488482 531 / 578 The train loss 0.264745447016 532 / 578 The train loss 0.264784871842 533 / 578 The train loss 0.264713477053 534 / 578 The train loss 0.264464431135 535 / 578 The train loss 0.264126362182 536 / 578 The train loss 0.263765178437 537 / 578 The train loss 0.263529065889 538 / 578 The train loss 0.263442086779 539 / 578 The train loss 0.263129524745 540 / 578 The train loss 0.262720279102 541 / 578 The train loss 0.262500971924 542 / 578 The train loss 0.262380618984 543 / 578 The train loss 0.262043836048 544 / 578 The train loss 0.261742725595 545 / 578 The train loss 0.261751340846 546 / 578 The train loss 0.26287914426 547 / 578 The train loss 0.264369569171 548 / 578 The train loss 0.266997616993 549 / 578 The train loss 0.268220471673 550 / 578 The train loss 0.268670710227 551 / 578 The train loss 0.269404512795 552 / 578 The train loss 0.269951977818 553 / 578 The train loss 0.270430174496 554 / 578 The train loss 0.27074248767 555 / 578 The train loss 0.270861770182 556 / 578 The train loss 0.271193297979 557 / 578 The train loss 0.271941708822 558 / 578 The train loss 0.272593296246 559 / 578 The train loss 0.272966192702 560 / 578 The train loss 0.273374647497 561 / 578 The train loss 0.273387907033 562 / 578 The train loss 0.273050507428 563 / 578 The train loss 0.273061139569 564 / 578 The train loss 0.273685359597 565 / 578 The train loss 0.27384814828 566 / 578 The train loss 0.273692480163 567 / 578 The train loss 0.273359850019 568 / 578 The train loss 0.273066324008 569 / 578 The train loss 0.272748647415 570 / 578 The train loss 0.272343088899 571 / 578 The train loss 0.272113926803 572 / 578 The train loss 0.271770011116 573 / 578 The train loss 0.271674724361 574 / 578 The train loss 0.271500290975 575 / 578 The train loss 0.271181124387 576 / 578 The train loss 0.271093949304 577 / 578 The train loss 0.270969134828 578 / 578 The train loss 0.270820517174 Starting epoch 16 Validation: 1 / 30 The valid loss 0.417902141809 2 / 30 The valid loss 0.401053279638 3 / 30 The valid loss 0.364517331123 4 / 30 The valid loss 0.292307499796 5 / 30 The valid loss 0.246447524428 6 / 30 The valid loss 0.258626726766 7 / 30 The valid loss 0.345412107451 8 / 30 The valid loss 0.475403806195 9 / 30 The valid loss 0.609615026249 10 / 30 The valid loss 0.680645863712 11 / 30 The valid loss 0.707709744573 12 / 30 The valid loss 0.72032796219 13 / 30 The valid loss 0.727850743211 14 / 30 The valid loss 0.733771012298 15 / 30 The valid loss 0.723980640372 16 / 30 The valid loss 0.721818624996 17 / 30 The valid loss 0.736480315819 18 / 30 The valid loss 0.747327230871 19 / 30 The valid loss 0.748518535181 20 / 30 The valid loss 0.745627453178 21 / 30 The valid loss 0.729933114988 22 / 30 The valid loss 0.71329885518 23 / 30 The valid loss 0.699327437126 24 / 30 The valid loss 0.687945139284 25 / 30 The valid loss 0.67373559773 26 / 30 The valid loss 0.663457393073 27 / 30 The valid loss 0.649255159277 28 / 30 The valid loss 0.632549384343 29 / 30 The valid loss 0.617529968249 30 / 30 The valid loss 0.606414295733 Validation MSE(val_loss): 12.8607292302 Test MSE(test_loss): 2.41759893091 Training 1 / 578 The train loss 0.242021888494 2 / 578 The train loss 0.14642803371 3 / 578 The train loss 0.131260876854 4 / 578 The train loss 0.146088775247 5 / 578 The train loss 0.140045696497 6 / 578 The train loss 0.138266103963 7 / 578 The train loss 0.141008472868 8 / 578 The train loss 0.132472227328 9 / 578 The train loss 0.123641769919 10 / 578 The train loss 0.117567428201 11 / 578 The train loss 0.111605083739 12 / 578 The train loss 0.106647755019 13 / 578 The train loss 0.102486381737 14 / 578 The train loss 0.10340030651 15 / 578 The train loss 0.103573599954 16 / 578 The train loss 0.105502670165 17 / 578 The train loss 0.108974782421 18 / 578 The train loss 0.117430009362 19 / 578 The train loss 0.119847819601 20 / 578 The train loss 0.117780746892 21 / 578 The train loss 0.114336064351 22 / 578 The train loss 0.110545586033 23 / 578 The train loss 0.108059001601 24 / 578 The train loss 0.105928661845 25 / 578 The train loss 0.105763519332 26 / 578 The train loss 0.104341669366 27 / 578 The train loss 0.103530270841 28 / 578 The train loss 0.10184130811 29 / 578 The train loss 0.0992267567921 30 / 578 The train loss 0.0971698531881 31 / 578 The train loss 0.0949742721334 32 / 578 The train loss 0.0923912279541 33 / 578 The train loss 0.09032662258 34 / 578 The train loss 0.0888821416918 35 / 578 The train loss 0.0882449145828 36 / 578 The train loss 0.0913385264575 37 / 578 The train loss 0.0932336981232 38 / 578 The train loss 0.0934529069223 39 / 578 The train loss 0.0928905675044 40 / 578 The train loss 0.0919361276552 41 / 578 The train loss 0.0900651003666 42 / 578 The train loss 0.0885279485291 43 / 578 The train loss 0.0874400662778 44 / 578 The train loss 0.0860183209138 45 / 578 The train loss 0.0849098507522 46 / 578 The train loss 0.0843454123799 47 / 578 The train loss 0.0836243772721 48 / 578 The train loss 0.0838957497811 49 / 578 The train loss 0.0834106403612 50 / 578 The train loss 0.0829098003171 51 / 578 The train loss 0.082137301619 52 / 578 The train loss 0.082394518585 53 / 578 The train loss 0.0824839504225 54 / 578 The train loss 0.0818492300912 55 / 578 The train loss 0.081221235662 56 / 578 The train loss 0.0802316761276 57 / 578 The train loss 0.0801006819432 58 / 578 The train loss 0.0800484112978 59 / 578 The train loss 0.0810298823546 60 / 578 The train loss 0.0824903577411 61 / 578 The train loss 0.0870095066207 62 / 578 The train loss 0.0929819931757 63 / 578 The train loss 0.0976271927209 64 / 578 The train loss 0.100160147049 65 / 578 The train loss 0.101009002543 66 / 578 The train loss 0.10121829637 67 / 578 The train loss 0.100780089719 68 / 578 The train loss 0.100292456706 69 / 578 The train loss 0.0996543858335 70 / 578 The train loss 0.0990606386081 71 / 578 The train loss 0.0987372141672 72 / 578 The train loss 0.0976402538492 73 / 578 The train loss 0.0970239177831 74 / 578 The train loss 0.0987970412054 75 / 578 The train loss 0.102810359187 76 / 578 The train loss 0.106290931164 77 / 578 The train loss 0.10738880486 78 / 578 The train loss 0.108695674389 79 / 578 The train loss 0.111830752095 80 / 578 The train loss 0.113664628926 81 / 578 The train loss 0.115387543105 82 / 578 The train loss 0.116112195671 83 / 578 The train loss 0.117062631669 84 / 578 The train loss 0.11838344785 85 / 578 The train loss 0.119865163624 86 / 578 The train loss 0.120596346629 87 / 578 The train loss 0.120624631119 88 / 578 The train loss 0.120832581909 89 / 578 The train loss 0.124345859607 90 / 578 The train loss 0.1257531363 91 / 578 The train loss 0.124818218104 92 / 578 The train loss 0.124582772983 93 / 578 The train loss 0.124482796227 94 / 578 The train loss 0.124325882713 95 / 578 The train loss 0.124614071993 96 / 578 The train loss 0.124461493445 97 / 578 The train loss 0.124858524986 98 / 578 The train loss 0.125071720952 99 / 578 The train loss 0.12573559506 100 / 578 The train loss 0.12684719055 101 / 578 The train loss 0.128059472474 102 / 578 The train loss 0.12817618659 103 / 578 The train loss 0.128030402993 104 / 578 The train loss 0.128289073553 105 / 578 The train loss 0.127989866824 106 / 578 The train loss 0.128003171935 107 / 578 The train loss 0.128117080386 108 / 578 The train loss 0.127614731512 109 / 578 The train loss 0.126833355746 110 / 578 The train loss 0.126423281626 111 / 578 The train loss 0.126554241916 112 / 578 The train loss 0.127328116505 113 / 578 The train loss 0.12896609022 114 / 578 The train loss 0.130808051275 115 / 578 The train loss 0.132856828031 116 / 578 The train loss 0.13527521805 117 / 578 The train loss 0.137141447092 118 / 578 The train loss 0.139780287828 119 / 578 The train loss 0.142752251009 120 / 578 The train loss 0.14540021574 121 / 578 The train loss 0.148111680376 122 / 578 The train loss 0.151002628446 123 / 578 The train loss 0.156072034152 124 / 578 The train loss 0.163681050515 125 / 578 The train loss 0.170724810652 126 / 578 The train loss 0.176425576735 127 / 578 The train loss 0.182317083797 128 / 578 The train loss 0.188294907617 129 / 578 The train loss 0.194291602322 130 / 578 The train loss 0.200053073401 131 / 578 The train loss 0.204817694157 132 / 578 The train loss 0.206409272872 133 / 578 The train loss 0.206277354368 134 / 578 The train loss 0.206152163828 135 / 578 The train loss 0.20656638371 136 / 578 The train loss 0.207612137413 137 / 578 The train loss 0.208988984183 138 / 578 The train loss 0.209232031914 139 / 578 The train loss 0.209587747935 140 / 578 The train loss 0.209463470582 141 / 578 The train loss 0.20967184667 142 / 578 The train loss 0.209348109243 143 / 578 The train loss 0.208750145412 144 / 578 The train loss 0.208839191221 145 / 578 The train loss 0.208857038405 146 / 578 The train loss 0.209719180544 147 / 578 The train loss 0.210888376542 148 / 578 The train loss 0.211394542268 149 / 578 The train loss 0.212339495546 150 / 578 The train loss 0.212553484632 151 / 578 The train loss 0.211769455803 152 / 578 The train loss 0.212047242112 153 / 578 The train loss 0.211679517632 154 / 578 The train loss 0.211192856742 155 / 578 The train loss 0.211442185788 156 / 578 The train loss 0.2114450878 157 / 578 The train loss 0.211507286515 158 / 578 The train loss 0.211439930274 159 / 578 The train loss 0.210805663517 160 / 578 The train loss 0.210386136541 161 / 578 The train loss 0.209666608601 162 / 578 The train loss 0.20867176673 163 / 578 The train loss 0.207928064291 164 / 578 The train loss 0.207306126286 165 / 578 The train loss 0.206519876754 166 / 578 The train loss 0.205780597174 167 / 578 The train loss 0.204987138616 168 / 578 The train loss 0.207967615126 169 / 578 The train loss 0.210952502312 170 / 578 The train loss 0.212928185322 171 / 578 The train loss 0.21460885883 172 / 578 The train loss 0.215175770673 173 / 578 The train loss 0.217138133227 174 / 578 The train loss 0.21903517116 175 / 578 The train loss 0.220236511161 176 / 578 The train loss 0.221788287602 177 / 578 The train loss 0.222695951845 178 / 578 The train loss 0.222756846332 179 / 578 The train loss 0.222977822152 180 / 578 The train loss 0.224459690119 181 / 578 The train loss 0.224634245547 182 / 578 The train loss 0.22387847965 183 / 578 The train loss 0.223510023405 184 / 578 The train loss 0.22525103331 185 / 578 The train loss 0.228072554412 186 / 578 The train loss 0.228323565591 187 / 578 The train loss 0.228965951374 188 / 578 The train loss 0.232840115933 189 / 578 The train loss 0.237344484062 190 / 578 The train loss 0.242586474253 191 / 578 The train loss 0.24406370526 192 / 578 The train loss 0.245638320121 193 / 578 The train loss 0.247185530339 194 / 578 The train loss 0.248850549518 195 / 578 The train loss 0.249686769697 196 / 578 The train loss 0.252774478578 197 / 578 The train loss 0.254505014714 198 / 578 The train loss 0.25673714496 199 / 578 The train loss 0.263593516809 200 / 578 The train loss 0.270172116212 201 / 578 The train loss 0.274483557415 202 / 578 The train loss 0.278651990564 203 / 578 The train loss 0.281133258334 204 / 578 The train loss 0.282297842109 205 / 578 The train loss 0.28178056687 206 / 578 The train loss 0.281179486959 207 / 578 The train loss 0.280814544424 208 / 578 The train loss 0.280367353837 209 / 578 The train loss 0.280691172281 210 / 578 The train loss 0.280257416175 211 / 578 The train loss 0.279803392566 212 / 578 The train loss 0.279218268822 213 / 578 The train loss 0.278726970134 214 / 578 The train loss 0.278509642315 215 / 578 The train loss 0.278468173305 216 / 578 The train loss 0.278225990957 217 / 578 The train loss 0.278269197856 218 / 578 The train loss 0.278047934905 219 / 578 The train loss 0.277715630978 220 / 578 The train loss 0.277304761429 221 / 578 The train loss 0.276984691262 222 / 578 The train loss 0.276355244735 223 / 578 The train loss 0.276200943109 224 / 578 The train loss 0.276079886439 225 / 578 The train loss 0.276824009398 226 / 578 The train loss 0.277702093236 227 / 578 The train loss 0.278902717997 228 / 578 The train loss 0.280287386927 229 / 578 The train loss 0.280862002764 230 / 578 The train loss 0.281139783087 231 / 578 The train loss 0.281847538092 232 / 578 The train loss 0.282754570483 233 / 578 The train loss 0.283687494835 234 / 578 The train loss 0.284907169556 235 / 578 The train loss 0.285854954649 236 / 578 The train loss 0.285960222054 237 / 578 The train loss 0.286469830345 238 / 578 The train loss 0.288076556135 239 / 578 The train loss 0.288422371409 240 / 578 The train loss 0.288099496358 241 / 578 The train loss 0.287674415327 242 / 578 The train loss 0.28760447733 243 / 578 The train loss 0.287918433037 244 / 578 The train loss 0.288396981938 245 / 578 The train loss 0.289618978007 246 / 578 The train loss 0.290425445654 247 / 578 The train loss 0.289623333054 248 / 578 The train loss 0.289273524638 249 / 578 The train loss 0.288998786267 250 / 578 The train loss 0.289261277597 251 / 578 The train loss 0.28958220873 252 / 578 The train loss 0.290070281065 253 / 578 The train loss 0.290954536523 254 / 578 The train loss 0.293000349771 255 / 578 The train loss 0.295392806781 256 / 578 The train loss 0.298967214225 257 / 578 The train loss 0.302904434165 258 / 578 The train loss 0.305295741528 259 / 578 The train loss 0.305964603887 260 / 578 The train loss 0.306160791143 261 / 578 The train loss 0.306110785535 262 / 578 The train loss 0.306708398819 263 / 578 The train loss 0.30677537131 264 / 578 The train loss 0.306847595846 265 / 578 The train loss 0.30697850094 266 / 578 The train loss 0.307683491096 267 / 578 The train loss 0.307787687741 268 / 578 The train loss 0.307589351057 269 / 578 The train loss 0.308080953774 270 / 578 The train loss 0.308768157644 271 / 578 The train loss 0.309061787373 272 / 578 The train loss 0.308975176485 273 / 578 The train loss 0.308414861215 274 / 578 The train loss 0.307734403198 275 / 578 The train loss 0.307129764946 276 / 578 The train loss 0.306614079833 277 / 578 The train loss 0.305912441682 278 / 578 The train loss 0.305105563424 279 / 578 The train loss 0.304333744013 280 / 578 The train loss 0.304140433386 281 / 578 The train loss 0.303981493519 282 / 578 The train loss 0.303501004746 283 / 578 The train loss 0.304598013087 284 / 578 The train loss 0.305337401641 285 / 578 The train loss 0.305027285986 286 / 578 The train loss 0.304832748531 287 / 578 The train loss 0.304674454303 288 / 578 The train loss 0.304502313682 289 / 578 The train loss 0.304392632661 290 / 578 The train loss 0.304474676821 291 / 578 The train loss 0.304319037947 292 / 578 The train loss 0.304043465367 293 / 578 The train loss 0.303876526918 294 / 578 The train loss 0.303418665754 295 / 578 The train loss 0.303037892039 296 / 578 The train loss 0.302314961213 297 / 578 The train loss 0.301963378468 298 / 578 The train loss 0.301595216414 299 / 578 The train loss 0.301125316451 300 / 578 The train loss 0.301276018356 301 / 578 The train loss 0.301864098903 302 / 578 The train loss 0.302493182569 303 / 578 The train loss 0.30295595835 304 / 578 The train loss 0.30334178744 305 / 578 The train loss 0.303276410586 306 / 578 The train loss 0.30290364755 307 / 578 The train loss 0.302515362329 308 / 578 The train loss 0.301917002558 309 / 578 The train loss 0.301426768095 310 / 578 The train loss 0.301075845221 311 / 578 The train loss 0.302176176001 312 / 578 The train loss 0.303680359777 313 / 578 The train loss 0.305022430338 314 / 578 The train loss 0.306775014714 315 / 578 The train loss 0.307084805468 316 / 578 The train loss 0.306645851237 317 / 578 The train loss 0.306131473225 318 / 578 The train loss 0.305510806503 319 / 578 The train loss 0.3052716586 320 / 578 The train loss 0.304750101882 321 / 578 The train loss 0.303905296696 322 / 578 The train loss 0.303145923099 323 / 578 The train loss 0.302509798778 324 / 578 The train loss 0.30182271155 325 / 578 The train loss 0.301065242723 326 / 578 The train loss 0.300278115066 327 / 578 The train loss 0.29959607827 328 / 578 The train loss 0.298868166416 329 / 578 The train loss 0.298107676715 330 / 578 The train loss 0.297380487234 331 / 578 The train loss 0.296652295751 332 / 578 The train loss 0.295851198639 333 / 578 The train loss 0.295128941466 334 / 578 The train loss 0.294381054026 335 / 578 The train loss 0.293724711203 336 / 578 The train loss 0.293120091834 337 / 578 The train loss 0.292750669746 338 / 578 The train loss 0.292507398375 339 / 578 The train loss 0.292105093035 340 / 578 The train loss 0.29156866436 341 / 578 The train loss 0.291001169527 342 / 578 The train loss 0.290642543808 343 / 578 The train loss 0.29006297961 344 / 578 The train loss 0.289496038919 345 / 578 The train loss 0.288990168433 346 / 578 The train loss 0.288360072319 347 / 578 The train loss 0.287851391921 348 / 578 The train loss 0.287688917628 349 / 578 The train loss 0.287759007718 350 / 578 The train loss 0.287575030186 351 / 578 The train loss 0.287211388251 352 / 578 The train loss 0.286694156458 353 / 578 The train loss 0.286226349651 354 / 578 The train loss 0.286101248881 355 / 578 The train loss 0.286467506671 356 / 578 The train loss 0.286466303823 357 / 578 The train loss 0.286189519586 358 / 578 The train loss 0.285941090609 359 / 578 The train loss 0.285790317017 360 / 578 The train loss 0.285957792264 361 / 578 The train loss 0.286280953351 362 / 578 The train loss 0.286465352934 363 / 578 The train loss 0.286322819082 364 / 578 The train loss 0.28606108452 365 / 578 The train loss 0.285873434955 366 / 578 The train loss 0.285556621511 367 / 578 The train loss 0.285261425347 368 / 578 The train loss 0.284829041786 369 / 578 The train loss 0.284170318958 370 / 578 The train loss 0.283533908112 371 / 578 The train loss 0.282900799686 372 / 578 The train loss 0.282304669778 373 / 578 The train loss 0.281705171367 374 / 578 The train loss 0.281097146157 375 / 578 The train loss 0.280539878684 376 / 578 The train loss 0.279882722453 377 / 578 The train loss 0.279245536288 378 / 578 The train loss 0.278583650652 379 / 578 The train loss 0.278011112091 380 / 578 The train loss 0.277516104813 381 / 578 The train loss 0.277074718083 382 / 578 The train loss 0.276782834232 383 / 578 The train loss 0.276736663603 384 / 578 The train loss 0.276489117942 385 / 578 The train loss 0.276087330691 386 / 578 The train loss 0.27553165436 387 / 578 The train loss 0.275163325943 388 / 578 The train loss 0.274888015499 389 / 578 The train loss 0.274378097133 390 / 578 The train loss 0.273907753694 391 / 578 The train loss 0.273396438633 392 / 578 The train loss 0.272833388672 393 / 578 The train loss 0.272237996083 394 / 578 The train loss 0.271604176618 395 / 578 The train loss 0.271019612345 396 / 578 The train loss 0.270711013575 397 / 578 The train loss 0.270345799604 398 / 578 The train loss 0.269944181598 399 / 578 The train loss 0.269363678261 400 / 578 The train loss 0.26896474093 401 / 578 The train loss 0.26870639975 402 / 578 The train loss 0.26821021059 403 / 578 The train loss 0.267669675477 404 / 578 The train loss 0.26710113869 405 / 578 The train loss 0.266699340284 406 / 578 The train loss 0.266548072587 407 / 578 The train loss 0.266388095208 408 / 578 The train loss 0.265907660024 409 / 578 The train loss 0.26671789988 410 / 578 The train loss 0.268094885247 411 / 578 The train loss 0.269278445281 412 / 578 The train loss 0.270084742391 413 / 578 The train loss 0.270328350606 414 / 578 The train loss 0.271113758113 415 / 578 The train loss 0.271554684035 416 / 578 The train loss 0.271111895946 417 / 578 The train loss 0.270529434217 418 / 578 The train loss 0.270021273917 419 / 578 The train loss 0.26964483433 420 / 578 The train loss 0.269258119057 421 / 578 The train loss 0.268730496406 422 / 578 The train loss 0.268148363747 423 / 578 The train loss 0.267570794648 424 / 578 The train loss 0.267154915766 425 / 578 The train loss 0.266955761224 426 / 578 The train loss 0.266778218125 427 / 578 The train loss 0.266874448806 428 / 578 The train loss 0.267021209957 429 / 578 The train loss 0.266808766809 430 / 578 The train loss 0.266941399991 431 / 578 The train loss 0.267028168931 432 / 578 The train loss 0.266924245781 433 / 578 The train loss 0.267242416995 434 / 578 The train loss 0.268913855422 435 / 578 The train loss 0.270452770229 436 / 578 The train loss 0.271274342259 437 / 578 The train loss 0.271375719498 438 / 578 The train loss 0.271040855832 439 / 578 The train loss 0.270489211114 440 / 578 The train loss 0.269920036648 441 / 578 The train loss 0.269432629742 442 / 578 The train loss 0.268884187137 443 / 578 The train loss 0.268304968553 444 / 578 The train loss 0.267995124123 445 / 578 The train loss 0.268164266556 446 / 578 The train loss 0.268786792324 447 / 578 The train loss 0.269003583608 448 / 578 The train loss 0.269326988801 449 / 578 The train loss 0.26961459096 450 / 578 The train loss 0.269565518956 451 / 578 The train loss 0.269658933848 452 / 578 The train loss 0.270104831754 453 / 578 The train loss 0.270263914761 454 / 578 The train loss 0.270304342914 455 / 578 The train loss 0.27029271302 456 / 578 The train loss 0.269990056518 457 / 578 The train loss 0.269684732704 458 / 578 The train loss 0.269225961738 459 / 578 The train loss 0.26872579252 460 / 578 The train loss 0.268352795734 461 / 578 The train loss 0.267931266458 462 / 578 The train loss 0.267653741203 463 / 578 The train loss 0.268255125082 464 / 578 The train loss 0.268852833243 465 / 578 The train loss 0.269161312012 466 / 578 The train loss 0.268998340829 467 / 578 The train loss 0.2687223248 468 / 578 The train loss 0.268502578969 469 / 578 The train loss 0.268277743416 470 / 578 The train loss 0.268033052444 471 / 578 The train loss 0.267902947221 472 / 578 The train loss 0.267641917189 473 / 578 The train loss 0.267281527737 474 / 578 The train loss 0.266819147112 475 / 578 The train loss 0.266336308571 476 / 578 The train loss 0.265962914425 477 / 578 The train loss 0.265697223376 478 / 578 The train loss 0.265249950096 479 / 578 The train loss 0.264787179414 480 / 578 The train loss 0.264438237097 481 / 578 The train loss 0.264167263517 482 / 578 The train loss 0.263868706481 483 / 578 The train loss 0.263522869529 484 / 578 The train loss 0.263144618318 485 / 578 The train loss 0.26277260143 486 / 578 The train loss 0.26239245287 487 / 578 The train loss 0.262017685547 488 / 578 The train loss 0.261734876097 489 / 578 The train loss 0.261483346897 490 / 578 The train loss 0.261495777678 491 / 578 The train loss 0.261525364082 492 / 578 The train loss 0.261308939433 493 / 578 The train loss 0.261305425114 494 / 578 The train loss 0.261096623672 495 / 578 The train loss 0.26087609179 496 / 578 The train loss 0.260767720056 497 / 578 The train loss 0.260835007719 498 / 578 The train loss 0.260780224825 499 / 578 The train loss 0.260542402303 500 / 578 The train loss 0.260238750862 501 / 578 The train loss 0.259889781819 502 / 578 The train loss 0.259445914103 503 / 578 The train loss 0.259046296893 504 / 578 The train loss 0.258770175369 505 / 578 The train loss 0.258453812194 506 / 578 The train loss 0.258087925085 507 / 578 The train loss 0.257725815935 508 / 578 The train loss 0.257352101112 509 / 578 The train loss 0.256982163343 510 / 578 The train loss 0.256689189099 511 / 578 The train loss 0.256368116581 512 / 578 The train loss 0.256046465673 513 / 578 The train loss 0.255758355223 514 / 578 The train loss 0.25541337945 515 / 578 The train loss 0.255012450304 516 / 578 The train loss 0.254819895251 517 / 578 The train loss 0.254702164557 518 / 578 The train loss 0.254620612834 519 / 578 The train loss 0.254595665762 520 / 578 The train loss 0.254771772158 521 / 578 The train loss 0.254873735049 522 / 578 The train loss 0.254574357138 523 / 578 The train loss 0.254344353985 524 / 578 The train loss 0.254191164796 525 / 578 The train loss 0.253990180991 526 / 578 The train loss 0.253688989613 527 / 578 The train loss 0.253450923914 528 / 578 The train loss 0.253358872849 529 / 578 The train loss 0.25332494138 530 / 578 The train loss 0.25320185461 531 / 578 The train loss 0.253026829224 532 / 578 The train loss 0.253000728102 533 / 578 The train loss 0.252976473947 534 / 578 The train loss 0.252709110552 535 / 578 The train loss 0.252414878703 536 / 578 The train loss 0.252067163185 537 / 578 The train loss 0.251856939685 538 / 578 The train loss 0.251702853338 539 / 578 The train loss 0.251453573373 540 / 578 The train loss 0.251160830592 541 / 578 The train loss 0.250971641455 542 / 578 The train loss 0.250772339888 543 / 578 The train loss 0.250495532052 544 / 578 The train loss 0.250192396112 545 / 578 The train loss 0.250195233854 546 / 578 The train loss 0.251101256387 547 / 578 The train loss 0.252066861329 548 / 578 The train loss 0.254516189321 549 / 578 The train loss 0.255902124221 550 / 578 The train loss 0.256430259606 551 / 578 The train loss 0.256760145585 552 / 578 The train loss 0.256811306787 553 / 578 The train loss 0.256993479409 554 / 578 The train loss 0.257118985314 555 / 578 The train loss 0.256965632282 556 / 578 The train loss 0.256876912331 557 / 578 The train loss 0.256570631035 558 / 578 The train loss 0.256428457037 559 / 578 The train loss 0.256318199175 560 / 578 The train loss 0.256139482227 561 / 578 The train loss 0.256094353976 562 / 578 The train loss 0.256044198333 563 / 578 The train loss 0.255923686494 564 / 578 The train loss 0.255663250449 565 / 578 The train loss 0.255303823185 566 / 578 The train loss 0.254959687579 567 / 578 The train loss 0.254718272342 568 / 578 The train loss 0.254476439249 569 / 578 The train loss 0.25422181112 570 / 578 The train loss 0.254076307164 571 / 578 The train loss 0.25373655839 572 / 578 The train loss 0.253535054463 573 / 578 The train loss 0.253169096156 574 / 578 The train loss 0.252892432022 575 / 578 The train loss 0.252585404912 576 / 578 The train loss 0.252867155665 577 / 578 The train loss 0.253496305635 578 / 578 The train loss 0.254236062256 Starting epoch 17 Validation: 1 / 30 The valid loss 0.456449568272 2 / 30 The valid loss 0.446170970798 3 / 30 The valid loss 0.404537717501 4 / 30 The valid loss 0.319109389558 5 / 30 The valid loss 0.265451798588 6 / 30 The valid loss 0.282125351951 7 / 30 The valid loss 0.376017126122 8 / 30 The valid loss 0.520632533822 9 / 30 The valid loss 0.669263759007 10 / 30 The valid loss 0.742962669209 11 / 30 The valid loss 0.794059438123 12 / 30 The valid loss 0.827145545743 13 / 30 The valid loss 0.847023614611 14 / 30 The valid loss 0.865216228845 15 / 30 The valid loss 0.858403443545 16 / 30 The valid loss 0.856127380626 17 / 30 The valid loss 0.873541768202 18 / 30 The valid loss 0.887138856161 19 / 30 The valid loss 0.891285914536 20 / 30 The valid loss 0.887925305404 21 / 30 The valid loss 0.869772672121 22 / 30 The valid loss 0.850275239653 23 / 30 The valid loss 0.834062012155 24 / 30 The valid loss 0.819509643596 25 / 30 The valid loss 0.801795058995 26 / 30 The valid loss 0.788870098069 27 / 30 The valid loss 0.769206544453 28 / 30 The valid loss 0.747561574115 29 / 30 The valid loss 0.729547003862 30 / 30 The valid loss 0.717483458544 Training 1 / 578 The train loss 0.403209894896 2 / 578 The train loss 0.266141474247 3 / 578 The train loss 0.247382019957 4 / 578 The train loss 0.278980705887 5 / 578 The train loss 0.262363809347 6 / 578 The train loss 0.234571763625 7 / 578 The train loss 0.217441980328 8 / 578 The train loss 0.199948566966 9 / 578 The train loss 0.181906540775 10 / 578 The train loss 0.167620457336 11 / 578 The train loss 0.155764657327 12 / 578 The train loss 0.146712987063 13 / 578 The train loss 0.139139375721 14 / 578 The train loss 0.135345420401 15 / 578 The train loss 0.1291503643 16 / 578 The train loss 0.127293118741 17 / 578 The train loss 0.130194914253 18 / 578 The train loss 0.130320875595 19 / 578 The train loss 0.130525737609 20 / 578 The train loss 0.127601933852 21 / 578 The train loss 0.123050117422 22 / 578 The train loss 0.119421970946 23 / 578 The train loss 0.116979156337 24 / 578 The train loss 0.114878016679 25 / 578 The train loss 0.113421343416 26 / 578 The train loss 0.111487225128 27 / 578 The train loss 0.109196164128 28 / 578 The train loss 0.106723963416 29 / 578 The train loss 0.104290663554 30 / 578 The train loss 0.10139455876 31 / 578 The train loss 0.0987560733432 32 / 578 The train loss 0.0964966252213 33 / 578 The train loss 0.0939633373566 34 / 578 The train loss 0.0919105576297 35 / 578 The train loss 0.0908870766472 36 / 578 The train loss 0.0916076564592 37 / 578 The train loss 0.091366679609 38 / 578 The train loss 0.0911269938132 39 / 578 The train loss 0.0900972176295 40 / 578 The train loss 0.0888790221186 41 / 578 The train loss 0.0875083229209 42 / 578 The train loss 0.0859165286778 43 / 578 The train loss 0.0843038349088 44 / 578 The train loss 0.0830054011577 45 / 578 The train loss 0.0817231540258 46 / 578 The train loss 0.0802784329078 47 / 578 The train loss 0.0793603444076 48 / 578 The train loss 0.0782005089374 49 / 578 The train loss 0.0776371499904 50 / 578 The train loss 0.0774124069698 51 / 578 The train loss 0.0774508200622 52 / 578 The train loss 0.078671107803 53 / 578 The train loss 0.0782562199945 54 / 578 The train loss 0.0773834053048 55 / 578 The train loss 0.0765451211313 56 / 578 The train loss 0.0754579370342 57 / 578 The train loss 0.0750898377839 58 / 578 The train loss 0.0754570307306 59 / 578 The train loss 0.0765034307773 60 / 578 The train loss 0.0787997699187 61 / 578 The train loss 0.0831822309735 62 / 578 The train loss 0.0893260640873 63 / 578 The train loss 0.0924514500306 64 / 578 The train loss 0.0940592045808 65 / 578 The train loss 0.0943636285714 66 / 578 The train loss 0.0940285773355 67 / 578 The train loss 0.0933928205145 68 / 578 The train loss 0.092971413903 69 / 578 The train loss 0.0920806686002 70 / 578 The train loss 0.0916851940033 71 / 578 The train loss 0.0926121051594 72 / 578 The train loss 0.0932744963017 73 / 578 The train loss 0.0939112358841 74 / 578 The train loss 0.0973755230095 75 / 578 The train loss 0.103030530227 76 / 578 The train loss 0.10634124594 77 / 578 The train loss 0.107453422899 78 / 578 The train loss 0.108609999996 79 / 578 The train loss 0.108970544207 80 / 578 The train loss 0.109697449778 81 / 578 The train loss 0.110710751139 82 / 578 The train loss 0.111767607684 83 / 578 The train loss 0.113436648835 84 / 578 The train loss 0.114896976356 85 / 578 The train loss 0.117698234155 86 / 578 The train loss 0.120472997265 87 / 578 The train loss 0.12280239714 88 / 578 The train loss 0.124872189534 89 / 578 The train loss 0.129653337474 90 / 578 The train loss 0.132712437554 91 / 578 The train loss 0.132659262036 92 / 578 The train loss 0.132445430555 93 / 578 The train loss 0.131758692774 94 / 578 The train loss 0.130973635727 95 / 578 The train loss 0.1303507883 96 / 578 The train loss 0.129682625382 97 / 578 The train loss 0.129026554573 98 / 578 The train loss 0.128546655378 99 / 578 The train loss 0.128824859152 100 / 578 The train loss 0.129414828764 101 / 578 The train loss 0.129424014645 102 / 578 The train loss 0.129039656855 103 / 578 The train loss 0.128648332284 104 / 578 The train loss 0.127759275752 105 / 578 The train loss 0.12695583883 106 / 578 The train loss 0.12639580573 107 / 578 The train loss 0.125927005041 108 / 578 The train loss 0.125573861283 109 / 578 The train loss 0.125479302523 110 / 578 The train loss 0.124950669071 111 / 578 The train loss 0.12470951996 112 / 578 The train loss 0.124561257502 113 / 578 The train loss 0.124780501519 114 / 578 The train loss 0.125063708583 115 / 578 The train loss 0.125590333059 116 / 578 The train loss 0.126469904725 117 / 578 The train loss 0.126945614409 118 / 578 The train loss 0.127465911691 119 / 578 The train loss 0.127522840321 120 / 578 The train loss 0.127776056505 121 / 578 The train loss 0.128279193966 122 / 578 The train loss 0.129323907662 123 / 578 The train loss 0.132032573564 124 / 578 The train loss 0.137139529328 125 / 578 The train loss 0.142921335079 126 / 578 The train loss 0.14790629475 127 / 578 The train loss 0.152749274989 128 / 578 The train loss 0.156694831188 129 / 578 The train loss 0.161578333518 130 / 578 The train loss 0.165886971906 131 / 578 The train loss 0.168192761182 132 / 578 The train loss 0.168887634712 133 / 578 The train loss 0.168658781972 134 / 578 The train loss 0.167940146644 135 / 578 The train loss 0.167494680022 136 / 578 The train loss 0.168397833737 137 / 578 The train loss 0.169033166463 138 / 578 The train loss 0.168813916524 139 / 578 The train loss 0.168871973253 140 / 578 The train loss 0.168955322654 141 / 578 The train loss 0.168968258155 142 / 578 The train loss 0.168624408996 143 / 578 The train loss 0.16867667221 144 / 578 The train loss 0.169391784746 145 / 578 The train loss 0.169627311344 146 / 578 The train loss 0.169905512434 147 / 578 The train loss 0.170576683651 148 / 578 The train loss 0.172121385597 149 / 578 The train loss 0.17449463747 150 / 578 The train loss 0.175256718329 151 / 578 The train loss 0.175134728219 152 / 578 The train loss 0.175534185739 153 / 578 The train loss 0.175867865026 154 / 578 The train loss 0.175379594165 155 / 578 The train loss 0.175529677667 156 / 578 The train loss 0.176324026546 157 / 578 The train loss 0.177164995776 158 / 578 The train loss 0.177330737185 159 / 578 The train loss 0.17740178308 160 / 578 The train loss 0.177111414756 161 / 578 The train loss 0.176607974504 162 / 578 The train loss 0.176306797795 163 / 578 The train loss 0.176314229643 164 / 578 The train loss 0.176008254227 165 / 578 The train loss 0.175339935026 166 / 578 The train loss 0.174891054973 167 / 578 The train loss 0.174779372106 168 / 578 The train loss 0.179405465173 169 / 578 The train loss 0.182207548567 170 / 578 The train loss 0.183066512716 171 / 578 The train loss 0.183391002345 172 / 578 The train loss 0.1833526012 173 / 578 The train loss 0.182961452026 174 / 578 The train loss 0.18283974489 175 / 578 The train loss 0.182625446644 176 / 578 The train loss 0.182569972308 177 / 578 The train loss 0.182052754692 178 / 578 The train loss 0.181767323823 179 / 578 The train loss 0.181730065292 180 / 578 The train loss 0.18190740424 181 / 578 The train loss 0.181702725245 182 / 578 The train loss 0.181197493189 183 / 578 The train loss 0.180871446586 184 / 578 The train loss 0.181426397631 185 / 578 The train loss 0.181592636802 186 / 578 The train loss 0.181192426376 187 / 578 The train loss 0.181720982844 188 / 578 The train loss 0.183417748209 189 / 578 The train loss 0.185429863459 190 / 578 The train loss 0.189396656447 191 / 578 The train loss 0.191824061031 192 / 578 The train loss 0.193957053072 193 / 578 The train loss 0.19760962775 194 / 578 The train loss 0.200015240671 195 / 578 The train loss 0.199829155001 196 / 578 The train loss 0.19924752371 197 / 578 The train loss 0.198755544024 198 / 578 The train loss 0.198414381128 199 / 578 The train loss 0.199119289041 200 / 578 The train loss 0.200209058695 201 / 578 The train loss 0.201071424294 202 / 578 The train loss 0.203190806797 203 / 578 The train loss 0.20433898133 204 / 578 The train loss 0.205104205415 205 / 578 The train loss 0.204781446829 206 / 578 The train loss 0.204658021649 207 / 578 The train loss 0.204703035009 208 / 578 The train loss 0.204525548629 209 / 578 The train loss 0.204759218142 210 / 578 The train loss 0.204541008977 211 / 578 The train loss 0.204218213614 212 / 578 The train loss 0.203788094455 213 / 578 The train loss 0.203548948412 214 / 578 The train loss 0.203720847532 215 / 578 The train loss 0.203614152262 216 / 578 The train loss 0.203818511515 217 / 578 The train loss 0.204478689212 218 / 578 The train loss 0.20501582354 219 / 578 The train loss 0.205620080934 220 / 578 The train loss 0.205633396406 221 / 578 The train loss 0.205038370847 222 / 578 The train loss 0.204752675371 223 / 578 The train loss 0.204883346898 224 / 578 The train loss 0.205073809771 225 / 578 The train loss 0.205204782275 226 / 578 The train loss 0.205688003339 227 / 578 The train loss 0.207076416064 228 / 578 The train loss 0.207605923269 229 / 578 The train loss 0.208242906807 230 / 578 The train loss 0.20889069508 231 / 578 The train loss 0.209311213123 232 / 578 The train loss 0.209801600592 233 / 578 The train loss 0.210850715817 234 / 578 The train loss 0.212209882059 235 / 578 The train loss 0.214905410549 236 / 578 The train loss 0.217927390161 237 / 578 The train loss 0.218960910835 238 / 578 The train loss 0.219357190313 239 / 578 The train loss 0.219568302492 240 / 578 The train loss 0.219822494736 241 / 578 The train loss 0.21999555621 242 / 578 The train loss 0.220958755456 243 / 578 The train loss 0.221425660648 244 / 578 The train loss 0.222049060098 245 / 578 The train loss 0.223546717046 246 / 578 The train loss 0.22554013877 247 / 578 The train loss 0.225123362431 248 / 578 The train loss 0.224778445246 249 / 578 The train loss 0.224516235251 250 / 578 The train loss 0.225177188087 251 / 578 The train loss 0.226316792486 252 / 578 The train loss 0.227162362139 253 / 578 The train loss 0.228565177711 254 / 578 The train loss 0.230350935835 255 / 578 The train loss 0.232026886754 256 / 578 The train loss 0.236027892493 257 / 578 The train loss 0.238274041596 258 / 578 The train loss 0.240257059466 259 / 578 The train loss 0.240944292256 260 / 578 The train loss 0.242166992266 261 / 578 The train loss 0.243037295685 262 / 578 The train loss 0.243435424408 263 / 578 The train loss 0.24363907825 264 / 578 The train loss 0.243925208328 265 / 578 The train loss 0.243941652427 266 / 578 The train loss 0.24380198753 267 / 578 The train loss 0.243612313974 268 / 578 The train loss 0.243039252421 269 / 578 The train loss 0.243217747385 270 / 578 The train loss 0.244818271686 271 / 578 The train loss 0.245931605286 272 / 578 The train loss 0.246295568413 273 / 578 The train loss 0.24643308682 274 / 578 The train loss 0.246472608587 275 / 578 The train loss 0.246304526427 276 / 578 The train loss 0.245986586076 277 / 578 The train loss 0.245393116811 278 / 578 The train loss 0.244790564623 279 / 578 The train loss 0.244285685871 280 / 578 The train loss 0.243776918406 281 / 578 The train loss 0.243543780243 282 / 578 The train loss 0.24311884423 283 / 578 The train loss 0.242639143655 284 / 578 The train loss 0.242262207959 285 / 578 The train loss 0.24188068712 286 / 578 The train loss 0.241632937073 287 / 578 The train loss 0.241266690003 288 / 578 The train loss 0.24102416944 289 / 578 The train loss 0.240614241321 290 / 578 The train loss 0.2401702639 291 / 578 The train loss 0.239758740627 292 / 578 The train loss 0.239450318857 293 / 578 The train loss 0.239175802094 294 / 578 The train loss 0.238781597659 295 / 578 The train loss 0.238301738456 296 / 578 The train loss 0.237884109553 297 / 578 The train loss 0.237319501227 298 / 578 The train loss 0.23690719029 299 / 578 The train loss 0.236362656687 300 / 578 The train loss 0.235954779601 301 / 578 The train loss 0.235790039431 302 / 578 The train loss 0.235954321591 303 / 578 The train loss 0.235885834798 304 / 578 The train loss 0.235884145065 305 / 578 The train loss 0.235987157475 306 / 578 The train loss 0.235573394398 307 / 578 The train loss 0.235324849927 308 / 578 The train loss 0.235011014861 309 / 578 The train loss 0.234675374488 310 / 578 The train loss 0.234659495965 311 / 578 The train loss 0.235561978698 312 / 578 The train loss 0.236965960719 313 / 578 The train loss 0.238205875181 314 / 578 The train loss 0.238992172362 315 / 578 The train loss 0.238628948577 316 / 578 The train loss 0.238186445856 317 / 578 The train loss 0.238158348199 318 / 578 The train loss 0.237886811311 319 / 578 The train loss 0.238096517657 320 / 578 The train loss 0.238423754941 321 / 578 The train loss 0.23804413856 322 / 578 The train loss 0.237630291446 323 / 578 The train loss 0.237167375967 324 / 578 The train loss 0.23664409172 325 / 578 The train loss 0.23611930546 326 / 578 The train loss 0.235558058132 327 / 578 The train loss 0.234940094107 328 / 578 The train loss 0.234341717351 329 / 578 The train loss 0.233718930874 330 / 578 The train loss 0.233224853729 331 / 578 The train loss 0.232690147968 332 / 578 The train loss 0.232072109135 333 / 578 The train loss 0.231540248964 334 / 578 The train loss 0.230970893839 335 / 578 The train loss 0.230449445526 336 / 578 The train loss 0.230011649943 337 / 578 The train loss 0.23002129579 338 / 578 The train loss 0.230585926364 339 / 578 The train loss 0.231148402756 340 / 578 The train loss 0.230917655789 341 / 578 The train loss 0.230375732028 342 / 578 The train loss 0.230067422311 343 / 578 The train loss 0.229660248588 344 / 578 The train loss 0.229204863302 345 / 578 The train loss 0.228822372565 346 / 578 The train loss 0.228469732322 347 / 578 The train loss 0.228158576894 348 / 578 The train loss 0.227966623437 349 / 578 The train loss 0.227980561995 350 / 578 The train loss 0.22803246757 351 / 578 The train loss 0.228291692082 352 / 578 The train loss 0.227996001435 353 / 578 The train loss 0.227579444238 354 / 578 The train loss 0.227383848566 355 / 578 The train loss 0.227582626121 356 / 578 The train loss 0.227675102832 357 / 578 The train loss 0.22740782544 358 / 578 The train loss 0.227406465987 359 / 578 The train loss 0.227299277427 360 / 578 The train loss 0.227452767488 361 / 578 The train loss 0.227733358135 362 / 578 The train loss 0.228415802381 363 / 578 The train loss 0.229462308509 364 / 578 The train loss 0.230042104878 365 / 578 The train loss 0.230051060054 366 / 578 The train loss 0.230436455654 367 / 578 The train loss 0.23015929707 368 / 578 The train loss 0.229845002023 369 / 578 The train loss 0.229403151701 370 / 578 The train loss 0.228913964928 371 / 578 The train loss 0.228472150995 372 / 578 The train loss 0.227989423972 373 / 578 The train loss 0.227718472034 374 / 578 The train loss 0.227665449958 375 / 578 The train loss 0.227327401889 376 / 578 The train loss 0.226795779851 377 / 578 The train loss 0.226290124296 378 / 578 The train loss 0.225850576074 379 / 578 The train loss 0.225410549292 380 / 578 The train loss 0.224869545826 381 / 578 The train loss 0.224552464906 382 / 578 The train loss 0.224324911504 383 / 578 The train loss 0.224003061635 384 / 578 The train loss 0.223598666433 385 / 578 The train loss 0.223161593585 386 / 578 The train loss 0.222768473099 387 / 578 The train loss 0.222276189061 388 / 578 The train loss 0.221832104721 389 / 578 The train loss 0.221409970566 390 / 578 The train loss 0.220919291525 391 / 578 The train loss 0.220549759164 392 / 578 The train loss 0.220064399557 393 / 578 The train loss 0.219613958721 394 / 578 The train loss 0.219092575774 395 / 578 The train loss 0.218613353011 396 / 578 The train loss 0.218316835509 397 / 578 The train loss 0.218098159023 398 / 578 The train loss 0.217857701599 399 / 578 The train loss 0.217430360277 400 / 578 The train loss 0.216991541747 401 / 578 The train loss 0.216563282277 402 / 578 The train loss 0.216115813357 403 / 578 The train loss 0.215729807596 404 / 578 The train loss 0.215604320701 405 / 578 The train loss 0.215591697429 406 / 578 The train loss 0.215752359037 407 / 578 The train loss 0.21563065504 408 / 578 The train loss 0.21518187862 409 / 578 The train loss 0.215837938294 410 / 578 The train loss 0.216973982658 411 / 578 The train loss 0.217410321234 412 / 578 The train loss 0.217752051311 413 / 578 The train loss 0.217697605434 414 / 578 The train loss 0.217868187701 415 / 578 The train loss 0.217934989909 416 / 578 The train loss 0.217791946047 417 / 578 The train loss 0.217697158257 418 / 578 The train loss 0.217693404448 419 / 578 The train loss 0.217870827166 420 / 578 The train loss 0.217752174694 421 / 578 The train loss 0.217361576676 422 / 578 The train loss 0.216909723231 423 / 578 The train loss 0.216504068272 424 / 578 The train loss 0.216128269665 425 / 578 The train loss 0.215908592499 426 / 578 The train loss 0.216180836211 427 / 578 The train loss 0.216412484905 428 / 578 The train loss 0.21643638017 429 / 578 The train loss 0.216528899931 430 / 578 The train loss 0.216974907262 431 / 578 The train loss 0.21716598103 432 / 578 The train loss 0.217252778146 433 / 578 The train loss 0.217707150646 434 / 578 The train loss 0.218323639716 435 / 578 The train loss 0.218733097268 436 / 578 The train loss 0.219482833415 437 / 578 The train loss 0.21939235992 438 / 578 The train loss 0.218995391138 439 / 578 The train loss 0.21860663949 440 / 578 The train loss 0.218332659874 441 / 578 The train loss 0.218012512049 442 / 578 The train loss 0.217687410414 443 / 578 The train loss 0.21738845658 444 / 578 The train loss 0.217244616464 445 / 578 The train loss 0.21774205736 446 / 578 The train loss 0.218370384664 447 / 578 The train loss 0.218964588287 448 / 578 The train loss 0.219033445529 449 / 578 The train loss 0.218996569645 450 / 578 The train loss 0.219016052747 451 / 578 The train loss 0.219173886029 452 / 578 The train loss 0.219213761354 453 / 578 The train loss 0.21893349594 454 / 578 The train loss 0.218573912993 455 / 578 The train loss 0.218211979098 456 / 578 The train loss 0.21791597583 457 / 578 The train loss 0.217701563606 458 / 578 The train loss 0.21737441762 459 / 578 The train loss 0.216965150768 460 / 578 The train loss 0.216554605465 461 / 578 The train loss 0.216184209691 462 / 578 The train loss 0.215892501633 463 / 578 The train loss 0.216842470145 464 / 578 The train loss 0.217542674431 465 / 578 The train loss 0.217688147521 466 / 578 The train loss 0.217527530262 467 / 578 The train loss 0.217400554582 468 / 578 The train loss 0.217270446963 469 / 578 The train loss 0.217197278802 470 / 578 The train loss 0.217150674703 471 / 578 The train loss 0.217098743394 472 / 578 The train loss 0.216944199977 473 / 578 The train loss 0.216798010645 474 / 578 The train loss 0.216508080521 475 / 578 The train loss 0.216240044571 476 / 578 The train loss 0.216165709886 477 / 578 The train loss 0.215967054049 478 / 578 The train loss 0.215692085037 479 / 578 The train loss 0.215344808418 480 / 578 The train loss 0.215002511044 481 / 578 The train loss 0.21476054378 482 / 578 The train loss 0.21446877176 483 / 578 The train loss 0.214145489629 484 / 578 The train loss 0.213796296832 485 / 578 The train loss 0.21347706287 486 / 578 The train loss 0.213210255734 487 / 578 The train loss 0.212928084834 488 / 578 The train loss 0.21259725744 489 / 578 The train loss 0.212361837059 490 / 578 The train loss 0.212314392624 491 / 578 The train loss 0.212288936716 492 / 578 The train loss 0.212276956685 493 / 578 The train loss 0.212324139129 494 / 578 The train loss 0.212165024235 495 / 578 The train loss 0.211999310386 496 / 578 The train loss 0.211936847245 497 / 578 The train loss 0.212046571389 498 / 578 The train loss 0.21208664269 499 / 578 The train loss 0.211982775993 500 / 578 The train loss 0.211726935415 501 / 578 The train loss 0.211377219844 502 / 578 The train loss 0.211010306203 503 / 578 The train loss 0.210764024346 504 / 578 The train loss 0.21054515082 505 / 578 The train loss 0.210242702146 506 / 578 The train loss 0.209991985577 507 / 578 The train loss 0.209738162809 508 / 578 The train loss 0.209457784222 509 / 578 The train loss 0.209131257991 510 / 578 The train loss 0.208825926006 511 / 578 The train loss 0.208536674378 512 / 578 The train loss 0.208195082565 513 / 578 The train loss 0.207862479309 514 / 578 The train loss 0.207488393152 515 / 578 The train loss 0.207233670551 516 / 578 The train loss 0.207005662275 517 / 578 The train loss 0.206868805778 518 / 578 The train loss 0.206726217759 519 / 578 The train loss 0.206532476805 520 / 578 The train loss 0.206497672109 521 / 578 The train loss 0.206470794241 522 / 578 The train loss 0.2061719979 523 / 578 The train loss 0.205904534947 524 / 578 The train loss 0.205674913543 525 / 578 The train loss 0.205580079607 526 / 578 The train loss 0.205438340509 527 / 578 The train loss 0.205353830803 528 / 578 The train loss 0.205321839526 529 / 578 The train loss 0.205369373063 530 / 578 The train loss 0.205380494353 531 / 578 The train loss 0.20548819395 532 / 578 The train loss 0.205765049311 533 / 578 The train loss 0.205914339692 534 / 578 The train loss 0.205846076177 535 / 578 The train loss 0.205695063316 536 / 578 The train loss 0.2054967975 537 / 578 The train loss 0.205353121957 538 / 578 The train loss 0.205173759678 539 / 578 The train loss 0.20494445511 540 / 578 The train loss 0.204656424313 541 / 578 The train loss 0.204487136018 542 / 578 The train loss 0.204465755342 543 / 578 The train loss 0.204249175387 544 / 578 The train loss 0.203966986664 545 / 578 The train loss 0.203959455022 546 / 578 The train loss 0.204493165455 547 / 578 The train loss 0.20504673482 548 / 578 The train loss 0.206668976234 549 / 578 The train loss 0.207284117224 550 / 578 The train loss 0.207522750769 551 / 578 The train loss 0.207556808007 552 / 578 The train loss 0.20762042831 553 / 578 The train loss 0.20786003103 554 / 578 The train loss 0.207838155304 555 / 578 The train loss 0.207686429667 556 / 578 The train loss 0.207632604626 557 / 578 The train loss 0.207563730009 558 / 578 The train loss 0.207412198975 559 / 578 The train loss 0.207246401137 560 / 578 The train loss 0.207016433737 561 / 578 The train loss 0.206695310358 562 / 578 The train loss 0.206468133765 563 / 578 The train loss 0.206436838841 564 / 578 The train loss 0.206176330142 565 / 578 The train loss 0.205848445778 566 / 578 The train loss 0.205541127439 567 / 578 The train loss 0.205242770113 568 / 578 The train loss 0.205002617735 569 / 578 The train loss 0.204812028812 570 / 578 The train loss 0.204597426312 571 / 578 The train loss 0.204480680963 572 / 578 The train loss 0.2043789004 573 / 578 The train loss 0.204124537219 574 / 578 The train loss 0.204024227072 575 / 578 The train loss 0.203935684939 576 / 578 The train loss 0.203661290801 577 / 578 The train loss 0.203580990048 578 / 578 The train loss 0.203440917767 Starting epoch 18 Validation: 1 / 30 The valid loss 0.37879756093 2 / 30 The valid loss 0.362038105726 3 / 30 The valid loss 0.33437760671 4 / 30 The valid loss 0.266196955927 5 / 30 The valid loss 0.224794021994 6 / 30 The valid loss 0.243225955094 7 / 30 The valid loss 0.334933317666 8 / 30 The valid loss 0.475273820106 9 / 30 The valid loss 0.618078856419 10 / 30 The valid loss 0.687568129972 11 / 30 The valid loss 0.719424737109 12 / 30 The valid loss 0.726316930416 13 / 30 The valid loss 0.725846888068 14 / 30 The valid loss 0.72324656296 15 / 30 The valid loss 0.706983952969 16 / 30 The valid loss 0.695825741859 17 / 30 The valid loss 0.69972593403 18 / 30 The valid loss 0.706842837441 19 / 30 The valid loss 0.709284846132 20 / 30 The valid loss 0.708847508766 21 / 30 The valid loss 0.693959939515 22 / 30 The valid loss 0.676660783758 23 / 30 The valid loss 0.66088001835 24 / 30 The valid loss 0.646734609734 25 / 30 The valid loss 0.630853565484 26 / 30 The valid loss 0.618880755196 27 / 30 The valid loss 0.603583323873 28 / 30 The valid loss 0.586145951679 29 / 30 The valid loss 0.570229788921 30 / 30 The valid loss 0.558606586481 Validation MSE(val_loss): 11.8468315061 Test MSE(test_loss): 1.99327794193 Training 1 / 578 The train loss 0.260988622904 2 / 578 The train loss 0.208772286773 3 / 578 The train loss 0.229863862197 4 / 578 The train loss 0.272019706666 5 / 578 The train loss 0.253463691473 6 / 578 The train loss 0.233172660073 7 / 578 The train loss 0.205627865025 8 / 578 The train loss 0.18275324977 9 / 578 The train loss 0.164566893544 10 / 578 The train loss 0.151523673162 11 / 578 The train loss 0.141294942999 12 / 578 The train loss 0.132897964368 13 / 578 The train loss 0.128526038275 14 / 578 The train loss 0.130694451609 15 / 578 The train loss 0.12963010172 16 / 578 The train loss 0.127959969454 17 / 578 The train loss 0.129419083981 18 / 578 The train loss 0.135126142038 19 / 578 The train loss 0.137014249438 20 / 578 The train loss 0.135805200413 21 / 578 The train loss 0.14154286789 22 / 578 The train loss 0.145226889375 23 / 578 The train loss 0.141981986878 24 / 578 The train loss 0.139674479452 25 / 578 The train loss 0.136024314314 26 / 578 The train loss 0.132494270945 27 / 578 The train loss 0.129440065749 28 / 578 The train loss 0.125437824455 29 / 578 The train loss 0.121775029163 30 / 578 The train loss 0.120030311371 31 / 578 The train loss 0.118131238847 32 / 578 The train loss 0.115326890897 33 / 578 The train loss 0.112634138007 34 / 578 The train loss 0.110163470511 35 / 578 The train loss 0.110032326515 36 / 578 The train loss 0.11278311008 37 / 578 The train loss 0.114821328203 38 / 578 The train loss 0.116873891044 39 / 578 The train loss 0.115938314929 40 / 578 The train loss 0.114370818343 41 / 578 The train loss 0.111964564316 42 / 578 The train loss 0.109664901337 43 / 578 The train loss 0.107743193374 44 / 578 The train loss 0.106010995001 45 / 578 The train loss 0.1045303668 46 / 578 The train loss 0.102825317248 47 / 578 The train loss 0.10091039796 48 / 578 The train loss 0.0993551158269 49 / 578 The train loss 0.0981144808917 50 / 578 The train loss 0.0969782320037 51 / 578 The train loss 0.0959449954258 52 / 578 The train loss 0.0955569744468 53 / 578 The train loss 0.0947600026387 54 / 578 The train loss 0.0942465586726 55 / 578 The train loss 0.0928907939995 56 / 578 The train loss 0.0915109684658 57 / 578 The train loss 0.0906399890225 58 / 578 The train loss 0.0898879084832 59 / 578 The train loss 0.0895131580626 60 / 578 The train loss 0.090740488392 61 / 578 The train loss 0.0930541283985 62 / 578 The train loss 0.0952997129981 63 / 578 The train loss 0.0975031627283 64 / 578 The train loss 0.0984947003162 65 / 578 The train loss 0.0990454627201 66 / 578 The train loss 0.0991852896881 67 / 578 The train loss 0.0989487748732 68 / 578 The train loss 0.098485940451 69 / 578 The train loss 0.0976210303902 70 / 578 The train loss 0.0970141874227 71 / 578 The train loss 0.0968983563083 72 / 578 The train loss 0.0963171556359 73 / 578 The train loss 0.096227855243 74 / 578 The train loss 0.101399858595 75 / 578 The train loss 0.109777278441 76 / 578 The train loss 0.113469251223 77 / 578 The train loss 0.114950941501 78 / 578 The train loss 0.116360023367 79 / 578 The train loss 0.116843154477 80 / 578 The train loss 0.11743597904 81 / 578 The train loss 0.117386415543 82 / 578 The train loss 0.118817189424 83 / 578 The train loss 0.119242594847 84 / 578 The train loss 0.119198371024 85 / 578 The train loss 0.119257055081 86 / 578 The train loss 0.119397943742 87 / 578 The train loss 0.11873571094 88 / 578 The train loss 0.118220638812 89 / 578 The train loss 0.120828871636 90 / 578 The train loss 0.121937423769 91 / 578 The train loss 0.121108508607 92 / 578 The train loss 0.120082035205 93 / 578 The train loss 0.119141408062 94 / 578 The train loss 0.118353931382 95 / 578 The train loss 0.118164298517 96 / 578 The train loss 0.117762260556 97 / 578 The train loss 0.11749763805 98 / 578 The train loss 0.117275028157 99 / 578 The train loss 0.117815700548 100 / 578 The train loss 0.119573445423 101 / 578 The train loss 0.120435067877 102 / 578 The train loss 0.120708007247 103 / 578 The train loss 0.120494155439 104 / 578 The train loss 0.120059797352 105 / 578 The train loss 0.119558470829 106 / 578 The train loss 0.119012619026 107 / 578 The train loss 0.118730344427 108 / 578 The train loss 0.11847669927 109 / 578 The train loss 0.117906957207 110 / 578 The train loss 0.1174963759 111 / 578 The train loss 0.117648480334 112 / 578 The train loss 0.117678496812 113 / 578 The train loss 0.117984020827 114 / 578 The train loss 0.118411346183 115 / 578 The train loss 0.118598277502 116 / 578 The train loss 0.119207637537 117 / 578 The train loss 0.119682437931 118 / 578 The train loss 0.120441840827 119 / 578 The train loss 0.120900354677 120 / 578 The train loss 0.121693081742 121 / 578 The train loss 0.122921193765 122 / 578 The train loss 0.124193712176 123 / 578 The train loss 0.126260873589 124 / 578 The train loss 0.130509741687 125 / 578 The train loss 0.134505992733 126 / 578 The train loss 0.138116758812 127 / 578 The train loss 0.141745753022 128 / 578 The train loss 0.146144992519 129 / 578 The train loss 0.150495260109 130 / 578 The train loss 0.15573561956 131 / 578 The train loss 0.159812220941 132 / 578 The train loss 0.162297904569 133 / 578 The train loss 0.162830117467 134 / 578 The train loss 0.162567601908 135 / 578 The train loss 0.163041210471 136 / 578 The train loss 0.16354843894 137 / 578 The train loss 0.163896659287 138 / 578 The train loss 0.163747800183 139 / 578 The train loss 0.163673678393 140 / 578 The train loss 0.163038867359 141 / 578 The train loss 0.162223692977 142 / 578 The train loss 0.161515426802 143 / 578 The train loss 0.161434567599 144 / 578 The train loss 0.162040297065 145 / 578 The train loss 0.16234994584 146 / 578 The train loss 0.16211747278 147 / 578 The train loss 0.162496615043 148 / 578 The train loss 0.162976651597 149 / 578 The train loss 0.164142641464 150 / 578 The train loss 0.164939362624 151 / 578 The train loss 0.164477575268 152 / 578 The train loss 0.164007531595 153 / 578 The train loss 0.163990708756 154 / 578 The train loss 0.163708146535 155 / 578 The train loss 0.163148684962 156 / 578 The train loss 0.163398891705 157 / 578 The train loss 0.163217790395 158 / 578 The train loss 0.162930026668 159 / 578 The train loss 0.162251713018 160 / 578 The train loss 0.161586810573 161 / 578 The train loss 0.161015224631 162 / 578 The train loss 0.160533881499 163 / 578 The train loss 0.160355234182 164 / 578 The train loss 0.160299495606 165 / 578 The train loss 0.160149242046 166 / 578 The train loss 0.159803589217 167 / 578 The train loss 0.160085472847 168 / 578 The train loss 0.162968151598 169 / 578 The train loss 0.163497452172 170 / 578 The train loss 0.163085924193 171 / 578 The train loss 0.162540415896 172 / 578 The train loss 0.162095322701 173 / 578 The train loss 0.161922091365 174 / 578 The train loss 0.163513154321 175 / 578 The train loss 0.163558109674 176 / 578 The train loss 0.163634199496 177 / 578 The train loss 0.163232418751 178 / 578 The train loss 0.162949220659 179 / 578 The train loss 0.16263590609 180 / 578 The train loss 0.162241557256 181 / 578 The train loss 0.161868624487 182 / 578 The train loss 0.161345871201 183 / 578 The train loss 0.161018570687 184 / 578 The train loss 0.161768840317 185 / 578 The train loss 0.162053753733 186 / 578 The train loss 0.161716139219 187 / 578 The train loss 0.161555510886 188 / 578 The train loss 0.161738632777 189 / 578 The train loss 0.161843981155 190 / 578 The train loss 0.164026475046 191 / 578 The train loss 0.166927094952 192 / 578 The train loss 0.168939742973 193 / 578 The train loss 0.170687782733 194 / 578 The train loss 0.172999142155 195 / 578 The train loss 0.173393815388 196 / 578 The train loss 0.172813967303 197 / 578 The train loss 0.172451905315 198 / 578 The train loss 0.172346802348 199 / 578 The train loss 0.173627485708 200 / 578 The train loss 0.17551505161 201 / 578 The train loss 0.176169242557 202 / 578 The train loss 0.17826706449 203 / 578 The train loss 0.179721183501 204 / 578 The train loss 0.180234215839 205 / 578 The train loss 0.179867055771 206 / 578 The train loss 0.179850421172 207 / 578 The train loss 0.179824361744 208 / 578 The train loss 0.179629049886 209 / 578 The train loss 0.180161096525 210 / 578 The train loss 0.179926570234 211 / 578 The train loss 0.179626369624 212 / 578 The train loss 0.179191222568 213 / 578 The train loss 0.178966878634 214 / 578 The train loss 0.178888698513 215 / 578 The train loss 0.179045181128 216 / 578 The train loss 0.179478018696 217 / 578 The train loss 0.179562912537 218 / 578 The train loss 0.179434018923 219 / 578 The train loss 0.17939479123 220 / 578 The train loss 0.179201450516 221 / 578 The train loss 0.178989310386 222 / 578 The train loss 0.178665471989 223 / 578 The train loss 0.178483851099 224 / 578 The train loss 0.178446127114 225 / 578 The train loss 0.178817938579 226 / 578 The train loss 0.17947720665 227 / 578 The train loss 0.180429240693 228 / 578 The train loss 0.180966122774 229 / 578 The train loss 0.181408655708 230 / 578 The train loss 0.181718008945 231 / 578 The train loss 0.181962620269 232 / 578 The train loss 0.182226819544 233 / 578 The train loss 0.18292675581 234 / 578 The train loss 0.183982397322 235 / 578 The train loss 0.185478976254 236 / 578 The train loss 0.185416077986 237 / 578 The train loss 0.185484912651 238 / 578 The train loss 0.186055193628 239 / 578 The train loss 0.186558752827 240 / 578 The train loss 0.186865092391 241 / 578 The train loss 0.187006719886 242 / 578 The train loss 0.187341923198 243 / 578 The train loss 0.187996372653 244 / 578 The train loss 0.188643768075 245 / 578 The train loss 0.18962630509 246 / 578 The train loss 0.190238878627 247 / 578 The train loss 0.189884563489 248 / 578 The train loss 0.189462205767 249 / 578 The train loss 0.189125130944 250 / 578 The train loss 0.189462986123 251 / 578 The train loss 0.190917915924 252 / 578 The train loss 0.19202890875 253 / 578 The train loss 0.193072417519 254 / 578 The train loss 0.194417584845 255 / 578 The train loss 0.195877889102 256 / 578 The train loss 0.198201329225 257 / 578 The train loss 0.200610497593 258 / 578 The train loss 0.201788818662 259 / 578 The train loss 0.202531291492 260 / 578 The train loss 0.203146723038 261 / 578 The train loss 0.203704212485 262 / 578 The train loss 0.203844025969 263 / 578 The train loss 0.203487193357 264 / 578 The train loss 0.203323704004 265 / 578 The train loss 0.202838571605 266 / 578 The train loss 0.20228838706 267 / 578 The train loss 0.201913127805 268 / 578 The train loss 0.201971884661 269 / 578 The train loss 0.202059729679 270 / 578 The train loss 0.202393634076 271 / 578 The train loss 0.202461549562 272 / 578 The train loss 0.202449577268 273 / 578 The train loss 0.202202960641 274 / 578 The train loss 0.202042327592 275 / 578 The train loss 0.202079110372 276 / 578 The train loss 0.202136090926 277 / 578 The train loss 0.201907286585 278 / 578 The train loss 0.201535283028 279 / 578 The train loss 0.201272147648 280 / 578 The train loss 0.201677092068 281 / 578 The train loss 0.202284187831 282 / 578 The train loss 0.20258757008 283 / 578 The train loss 0.202571115266 284 / 578 The train loss 0.202616838512 285 / 578 The train loss 0.20281949473 286 / 578 The train loss 0.202542145909 287 / 578 The train loss 0.202239922436 288 / 578 The train loss 0.202371202264 289 / 578 The train loss 0.202389889532 290 / 578 The train loss 0.202452087926 291 / 578 The train loss 0.202440186986 292 / 578 The train loss 0.202416564871 293 / 578 The train loss 0.202199042463 294 / 578 The train loss 0.201867586049 295 / 578 The train loss 0.201531162287 296 / 578 The train loss 0.201332348157 297 / 578 The train loss 0.201078682319 298 / 578 The train loss 0.20065068392 299 / 578 The train loss 0.200395631984 300 / 578 The train loss 0.200641250843 301 / 578 The train loss 0.200943824693 302 / 578 The train loss 0.201600511771 303 / 578 The train loss 0.201856776762 304 / 578 The train loss 0.201808015114 305 / 578 The train loss 0.201610468826 306 / 578 The train loss 0.201330087285 307 / 578 The train loss 0.201049905583 308 / 578 The train loss 0.200782592696 309 / 578 The train loss 0.200844453554 310 / 578 The train loss 0.200929150163 311 / 578 The train loss 0.201720219701 312 / 578 The train loss 0.202907131636 313 / 578 The train loss 0.203965091462 314 / 578 The train loss 0.205172938701 315 / 578 The train loss 0.205274773623 316 / 578 The train loss 0.204969177267 317 / 578 The train loss 0.204528951809 318 / 578 The train loss 0.204066096843 319 / 578 The train loss 0.203633027972 320 / 578 The train loss 0.203209068856 321 / 578 The train loss 0.20316944018 322 / 578 The train loss 0.202900659082 323 / 578 The train loss 0.202669647684 324 / 578 The train loss 0.202359612568 325 / 578 The train loss 0.20200575301 326 / 578 The train loss 0.20179232909 327 / 578 The train loss 0.201603103164 328 / 578 The train loss 0.201203827425 329 / 578 The train loss 0.200764428901 330 / 578 The train loss 0.200471838245 331 / 578 The train loss 0.199985477879 332 / 578 The train loss 0.199597788321 333 / 578 The train loss 0.199299958059 334 / 578 The train loss 0.198881987234 335 / 578 The train loss 0.19872867011 336 / 578 The train loss 0.198463258952 337 / 578 The train loss 0.198126755917 338 / 578 The train loss 0.197799840024 339 / 578 The train loss 0.197407659811 340 / 578 The train loss 0.19716691105 341 / 578 The train loss 0.197156363587 342 / 578 The train loss 0.197087571654 343 / 578 The train loss 0.196811319937 344 / 578 The train loss 0.196659128309 345 / 578 The train loss 0.196521400906 346 / 578 The train loss 0.196294561693 347 / 578 The train loss 0.19628703317 348 / 578 The train loss 0.196328963418 349 / 578 The train loss 0.196361957377 350 / 578 The train loss 0.196214023421 351 / 578 The train loss 0.196149892508 352 / 578 The train loss 0.195907424731 353 / 578 The train loss 0.195733996164 354 / 578 The train loss 0.195670814856 355 / 578 The train loss 0.195745624263 356 / 578 The train loss 0.195709389349 357 / 578 The train loss 0.195542708684 358 / 578 The train loss 0.195321565814 359 / 578 The train loss 0.195081470443 360 / 578 The train loss 0.195129998647 361 / 578 The train loss 0.195216369222 362 / 578 The train loss 0.1950604555 363 / 578 The train loss 0.194872599411 364 / 578 The train loss 0.194812578362 365 / 578 The train loss 0.194966836618 366 / 578 The train loss 0.195841502253 367 / 578 The train loss 0.19600895254 368 / 578 The train loss 0.195869093718 369 / 578 The train loss 0.195579548795 370 / 578 The train loss 0.195284748276 371 / 578 The train loss 0.194872993642 372 / 578 The train loss 0.194442229804 373 / 578 The train loss 0.194145337564 374 / 578 The train loss 0.193761210538 375 / 578 The train loss 0.19333040572 376 / 578 The train loss 0.192848199728 377 / 578 The train loss 0.192443649028 378 / 578 The train loss 0.192036099877 379 / 578 The train loss 0.191563879009 380 / 578 The train loss 0.191157406985 381 / 578 The train loss 0.191023242975 382 / 578 The train loss 0.191235287559 383 / 578 The train loss 0.191238508859 384 / 578 The train loss 0.190884212735 385 / 578 The train loss 0.190511951126 386 / 578 The train loss 0.190163312292 387 / 578 The train loss 0.189819350545 388 / 578 The train loss 0.189380824225 389 / 578 The train loss 0.188938041225 390 / 578 The train loss 0.188521121834 391 / 578 The train loss 0.188144139069 392 / 578 The train loss 0.187738397672 393 / 578 The train loss 0.187309936618 394 / 578 The train loss 0.186920479072 395 / 578 The train loss 0.186567830115 396 / 578 The train loss 0.186194314085 397 / 578 The train loss 0.185844462571 398 / 578 The train loss 0.185445474713 399 / 578 The train loss 0.185119089701 400 / 578 The train loss 0.184924584683 401 / 578 The train loss 0.185033492558 402 / 578 The train loss 0.185603601367 403 / 578 The train loss 0.185824932165 404 / 578 The train loss 0.185806211287 405 / 578 The train loss 0.18570144951 406 / 578 The train loss 0.185645213066 407 / 578 The train loss 0.185426682325 408 / 578 The train loss 0.185099665459 409 / 578 The train loss 0.186351517757 410 / 578 The train loss 0.18802079502 411 / 578 The train loss 0.189090468026 412 / 578 The train loss 0.189980521527 413 / 578 The train loss 0.189964103118 414 / 578 The train loss 0.190187575215 415 / 578 The train loss 0.190034809224 416 / 578 The train loss 0.1896615976 417 / 578 The train loss 0.189244117641 418 / 578 The train loss 0.188938520555 419 / 578 The train loss 0.188756825113 420 / 578 The train loss 0.188612676598 421 / 578 The train loss 0.188298105831 422 / 578 The train loss 0.187901664887 423 / 578 The train loss 0.187581050965 424 / 578 The train loss 0.187241571991 425 / 578 The train loss 0.186909463484 426 / 578 The train loss 0.186710389069 427 / 578 The train loss 0.186470538456 428 / 578 The train loss 0.186143408677 429 / 578 The train loss 0.185863622505 430 / 578 The train loss 0.185653819508 431 / 578 The train loss 0.185579164139 432 / 578 The train loss 0.185490430246 433 / 578 The train loss 0.185789362542 434 / 578 The train loss 0.186510228771 435 / 578 The train loss 0.187066662397 436 / 578 The train loss 0.187588291516 437 / 578 The train loss 0.18752198697 438 / 578 The train loss 0.187288572697 439 / 578 The train loss 0.186966004737 440 / 578 The train loss 0.1866865725 441 / 578 The train loss 0.186495280607 442 / 578 The train loss 0.18650165384 443 / 578 The train loss 0.186352668746 444 / 578 The train loss 0.18600514907 445 / 578 The train loss 0.1857294821 446 / 578 The train loss 0.185710956263 447 / 578 The train loss 0.185523967948 448 / 578 The train loss 0.185340697278 449 / 578 The train loss 0.185217553629 450 / 578 The train loss 0.185162024498 451 / 578 The train loss 0.18539201493 452 / 578 The train loss 0.185472251963 453 / 578 The train loss 0.185342434555 454 / 578 The train loss 0.185188628881 455 / 578 The train loss 0.185013645249 456 / 578 The train loss 0.184988021606 457 / 578 The train loss 0.184915684555 458 / 578 The train loss 0.184685100553 459 / 578 The train loss 0.184467797933 460 / 578 The train loss 0.184264509331 461 / 578 The train loss 0.183934030477 462 / 578 The train loss 0.183630379884 463 / 578 The train loss 0.184546961246 464 / 578 The train loss 0.184763745878 465 / 578 The train loss 0.18458610314 466 / 578 The train loss 0.184338578299 467 / 578 The train loss 0.184011899653 468 / 578 The train loss 0.183788654832 469 / 578 The train loss 0.183685712214 470 / 578 The train loss 0.183509516399 471 / 578 The train loss 0.183317505693 472 / 578 The train loss 0.183137248097 473 / 578 The train loss 0.182917326547 474 / 578 The train loss 0.182665190899 475 / 578 The train loss 0.182321828925 476 / 578 The train loss 0.182100786201 477 / 578 The train loss 0.181926046035 478 / 578 The train loss 0.181623980738 479 / 578 The train loss 0.181298719544 480 / 578 The train loss 0.180961638437 481 / 578 The train loss 0.180646077538 482 / 578 The train loss 0.180358799828 483 / 578 The train loss 0.180082150934 484 / 578 The train loss 0.17984141496 485 / 578 The train loss 0.179568631459 486 / 578 The train loss 0.179306295516 487 / 578 The train loss 0.179094115945 488 / 578 The train loss 0.178843947597 489 / 578 The train loss 0.178626007475 490 / 578 The train loss 0.178532800455 491 / 578 The train loss 0.178525569251 492 / 578 The train loss 0.178519203392 493 / 578 The train loss 0.178492390099 494 / 578 The train loss 0.178326317878 495 / 578 The train loss 0.178148152442 496 / 578 The train loss 0.178065337823 497 / 578 The train loss 0.178122940389 498 / 578 The train loss 0.178023363912 499 / 578 The train loss 0.17782543641 500 / 578 The train loss 0.177559744623 501 / 578 The train loss 0.177299152253 502 / 578 The train loss 0.177039683433 503 / 578 The train loss 0.176826191163 504 / 578 The train loss 0.176609651219 505 / 578 The train loss 0.176415721618 506 / 578 The train loss 0.176212179305 507 / 578 The train loss 0.176006574988 508 / 578 The train loss 0.175790678171 509 / 578 The train loss 0.175531217148 510 / 578 The train loss 0.175270581308 511 / 578 The train loss 0.175012515078 512 / 578 The train loss 0.174729581016 513 / 578 The train loss 0.174424227759 514 / 578 The train loss 0.174114120138 515 / 578 The train loss 0.173817389719 516 / 578 The train loss 0.173706390029 517 / 578 The train loss 0.17372040068 518 / 578 The train loss 0.173706008668 519 / 578 The train loss 0.173532027392 520 / 578 The train loss 0.173342701384 521 / 578 The train loss 0.173182198381 522 / 578 The train loss 0.172905539102 523 / 578 The train loss 0.172675559671 524 / 578 The train loss 0.172536426075 525 / 578 The train loss 0.172318650094 526 / 578 The train loss 0.172116997924 527 / 578 The train loss 0.172006192827 528 / 578 The train loss 0.17184185799 529 / 578 The train loss 0.171745222822 530 / 578 The train loss 0.171654246403 531 / 578 The train loss 0.171548040889 532 / 578 The train loss 0.1714996041 533 / 578 The train loss 0.171380869618 534 / 578 The train loss 0.1712650049 535 / 578 The train loss 0.171145574101 536 / 578 The train loss 0.170980395756 537 / 578 The train loss 0.17083830992 538 / 578 The train loss 0.17085637783 539 / 578 The train loss 0.170851724316 540 / 578 The train loss 0.170687428303 541 / 578 The train loss 0.170626901923 542 / 578 The train loss 0.170644814808 543 / 578 The train loss 0.170500338894 544 / 578 The train loss 0.17026069572 545 / 578 The train loss 0.170050557293 546 / 578 The train loss 0.170227315638 547 / 578 The train loss 0.1705330714 548 / 578 The train loss 0.171602820848 549 / 578 The train loss 0.172191936971 550 / 578 The train loss 0.172385062325 551 / 578 The train loss 0.172323294234 552 / 578 The train loss 0.172143408023 553 / 578 The train loss 0.171931936326 554 / 578 The train loss 0.171721262642 555 / 578 The train loss 0.171627562587 556 / 578 The train loss 0.171674400427 557 / 578 The train loss 0.171551652901 558 / 578 The train loss 0.171444814272 559 / 578 The train loss 0.171244721993 560 / 578 The train loss 0.171045686956 561 / 578 The train loss 0.170840993192 562 / 578 The train loss 0.170588472401 563 / 578 The train loss 0.170459080097 564 / 578 The train loss 0.170213128698 565 / 578 The train loss 0.170049703741 566 / 578 The train loss 0.169919214879 567 / 578 The train loss 0.169887607927 568 / 578 The train loss 0.169825940904 569 / 578 The train loss 0.169763280134 570 / 578 The train loss 0.169740328062 571 / 578 The train loss 0.169528601062 572 / 578 The train loss 0.169305655076 573 / 578 The train loss 0.169224238872 574 / 578 The train loss 0.169089255407 575 / 578 The train loss 0.169225878741 576 / 578 The train loss 0.16898681973 577 / 578 The train loss 0.168725681457 578 / 578 The train loss 0.168475877777 Starting epoch 19 Validation: 1 / 30 The valid loss 0.393577337265 2 / 30 The valid loss 0.397113829851 3 / 30 The valid loss 0.37603037556 4 / 30 The valid loss 0.303563499823 5 / 30 The valid loss 0.248843989521 6 / 30 The valid loss 0.244785980011 7 / 30 The valid loss 0.30897199735 8 / 30 The valid loss 0.424622487742 9 / 30 The valid loss 0.550123291297 10 / 30 The valid loss 0.609582779184 11 / 30 The valid loss 0.669435271147 12 / 30 The valid loss 0.709297605169 13 / 30 The valid loss 0.736754351224 14 / 30 The valid loss 0.75874156425 15 / 30 The valid loss 0.754378909121 16 / 30 The valid loss 0.752829356352 17 / 30 The valid loss 0.767296344261 18 / 30 The valid loss 0.778991158017 19 / 30 The valid loss 0.788201860887 20 / 30 The valid loss 0.792214180715 21 / 30 The valid loss 0.780035610887 22 / 30 The valid loss 0.764398730783 23 / 30 The valid loss 0.751558160166 24 / 30 The valid loss 0.740251342611 25 / 30 The valid loss 0.723737756163 26 / 30 The valid loss 0.710944385626 27 / 30 The valid loss 0.693705798989 28 / 30 The valid loss 0.673960500663 29 / 30 The valid loss 0.655496347182 30 / 30 The valid loss 0.640987485275
从上面可以看出，测试集表现最好的为：
Starting epoch 18
Validation MSE(val_loss): 11.8468315061
Test MSE(test_loss): 1.99327794193
分析：CNN+RNN seq2seq model表现不错，但反复实验几次后，发现模型很容易过拟合，表现很不稳定。究其原因还是训练数据太少，作为选做部分，将留待后面进一步优化。
Part 4: 选择最佳模型并生成结果视频
说明：
- 不使用选做部分CNN+RNN seq2seq model生成测试视频，留待后续完成
1.使用时间序列图对比各模型
In [7]:
### <1> NVIDIA end-to-end model：Benchmark (基准模型)
# 载入模型
from keras.models import load_model
import preprocess_data
import matplotlib.pyplot as plt
# 载入模型
model = load_model('./models/nvidia_model.h5')
#载入测试数据
test_imgs, test_wheels = preprocess_data.load_data('test')
test_imgs = preprocess_data.nomorlize_image(test_imgs)
#用模型生成预测转向角度
predicted_wheels = model.predict(test_imgs, batch_size=128, verbose=0)
#生成时间序列图
plt.figure
plt.title('Benchmark NVIDIA end-to-end model')
plt.plot(predicted_wheels)
plt.plot(test_wheels)
plt.ylabel('Steering angle', fontsize=11)
plt.xlabel('Frame counts', fontsize=11)
plt.legend(['Predicted Wheels', 'Ground Truth Wheels'], loc='upper right')
plt.xlim((0,2700))
plt.grid()
plt.savefig('./images/img/1_benchmark.png', dpi=300)
plt.show()
load data start! The epoch10 mkv is processing loading data filished!
In [8]:
### <2> NVIDIA end-to-end model：Refined Model (改进模型)
# 载入模型
model = load_model('./models/nvidia_refined_model.h5')
#用模型生成预测转向角度
predicted_wheels = model.predict(test_imgs, batch_size=128, verbose=0)
#生成时间序列图
plt.figure
plt.title('Refined NVIDIA end-to-end Model')
plt.plot(predicted_wheels)
plt.plot(test_wheels)
plt.ylabel('Steering angle', fontsize=11)
plt.xlabel('Frame counts', fontsize=11)
plt.legend(['Predicted Wheels', 'Ground Truth Wheels'], loc='upper right')
plt.xlim((0,2700))
plt.grid()
plt.savefig('./images/img/2_refined.png', dpi=300)
plt.show()
In [9]:
### <3> NVIDIA end-to-end model：Refined Model + 转向角度数据增加
# 载入模型
model = load_model('./models/nvidia_ra_model.h5')
#用模型生成预测转向角度
predicted_wheels = model.predict(test_imgs, batch_size=128, verbose=0)
#生成时间序列图
plt.figure
plt.title('Refined NVIDIA end-to-end Plus data Model')
plt.plot(predicted_wheels)
plt.plot(test_wheels)
plt.ylabel('Steering angle', fontsize=11)
plt.xlabel('Frame counts', fontsize=11)
plt.legend(['Predicted Wheels', 'Ground Truth Wheels'], loc='upper right')
plt.xlim((0,2700))
plt.grid()
plt.savefig('./images/img/3_ra.png', dpi=300)
plt.show()
In [10]:
### <4> VGG16 + Nvidia model：notop + top2 bolocks
# 载入模型
model = load_model('./models/vgg16_model.h5')
#用模型生成预测转向角度
predicted_wheels = model.predict(test_imgs, batch_size=128, verbose=0)
#生成时间序列图
plt.figure
plt.title('VGG16 + Nvidia model notop + top2 bolocks')
plt.plot(predicted_wheels)
plt.plot(test_wheels)
plt.ylabel('Steering angle', fontsize=11)
plt.xlabel('Frame counts', fontsize=11)
plt.legend(['Predicted Wheels', 'Ground Truth Wheels'], loc='upper right')
plt.xlim((0,2700))
plt.grid()
plt.savefig('./images/img/4_vgg16.png', dpi=300)
plt.show()
In [11]:
### <5> VGG16 + Nvidia model：notop + top2 bolocks + 转向角度数据增加
# 载入模型
model = load_model('./models/vgg16_add_model.h5')
#用模型生成预测转向角度
predicted_wheels = model.predict(test_imgs, batch_size=128, verbose=0)
#生成时间序列图
plt.figure
plt.title('VGG16 + Nvidia model notop + top2 bolocks plus data')
plt.plot(predicted_wheels)
plt.plot(test_wheels)
plt.ylabel('Steering angle', fontsize=11)
plt.xlabel('Frame counts', fontsize=11)
plt.legend(['Predicted Wheels', 'Ground Truth Wheels'], loc='upper right')
plt.xlim((0,2700))
plt.grid()
plt.savefig('./images/img/5_vgg16ad.png', dpi=300)
plt.show()
分析：从上面各模型的时间序列图可以看出，后两个VGG16+Nvidia model 明显比前三个Nvidia model输出的角度更平滑和贴合原始方向角度，同时从Part3各模型的'test loss'也可以看出后两个模型更出色。所以应该选用VGG16+Nvidia model，虽然'VGG16 + Nvidia model：notop + top2 bolocks'比'VGG16 + Nvidia model：notop + top2 bolocks + 转向角度数据增加'输出的'test loss'要小，但从图中可以看出，后者明显比前者的图形走势更接近于原始的数据走势，所以个人认为最佳的模型应该为最后一个模型'VGG16 + Nvidia model：notop + top2 bolocks + 转向角度数据增加'。
2.生成结果视频
策略：
- 1.修改./params.py文件中图片宽，高为'80'。
- 2.在./run.py文件中增加‘nomorlize_image()’函数，并在'img_pre_process()'中调用。
- 3.将./utils.py中'get_model'修改为载入vgg16_add_model.h5'和'vgg16_add_model.json'(405和406行)。运行'python run.py'命令生成视频
- 5.生成视频保存于./output文件夹中'epoch10_human_machine.mp4'文件
说明：首次运行时出现两个错误：1.需要下载'openh264-1.6.0-win64msvc.dll'文件放入项目文件根目录下（已解决）。2.修改./utils中'mkv_to_mp4()'函数适用于Windows 10系统，具体见./utils文件（已解决）
In [1]:
### 输出视频
from IPython.display import HTML
output = './output/epoch10_human_machine.mp4'
HTML("""
<video width="960" height="540" controls>
<source src="{0}">
</video>
""".format(output))
Out[1]:
In [ ]: